input
stringlengths 8.62k
18k
|
|---|
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Alexander in 1962, covered by the Beatles "Every Day I Have to Cry Some", covered by Ike and Tina Turner, produced by Phil Spector "Go Home Girl", recorded by Alexander in 1962 "In the Middle of it All", recorded by Alexander in 1987 "You Better Move On", recorded by Alexander in 1961, covered by the Hollies, the Rolling Stones, Billy "Crash" Craddock, Mink DeVille "Sally Sue Brown", recorded by Alexander in 1960, covered by Bob Dylan, Elvis Costello See also Muscle Shoals Sound Studio Soul music Country music References Bibliography Richard Younger; Get a Shot of Rhythm and Blues: The Arthur Alexander Story The University of Alabama Press (2000) External links Comprehensive Arthur Alexander Discography Arthur B. Alexander, Jr at FindaGrave Alabama Music Hall of Fame Interview With Richard Younger, "Get A Shot Of Rhythm And Blues: The Arthur Alexander Story" Category:1940 births Category:1993 deaths Category:Singers from Alabama Category:People from Sheffield, Alabama Category:American country singer-songwriters Category:American soul musicians Category:African-American singer-songwriters Category:Dot Records artists Category:Monument Records artists Category:20th-century American singers Category:African-American country musicians Category:Songwriters from Alabama Category:Country musicians from Alabama
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Shaheed A.H.M Qamaruzzaman Central Park and Zoo Shaheed A.H.M Qamaruzzaman Central Park and Zoo () is a public park as well as zoo in Rajshahi, Bangladesh. It is a popular recreation place for the citizens of the city. The park has a land area of 33 acres. It is named after A.H.M Qamaruzzaman, who was one of the national leader of Bangladesh. History During the British colonial period it was a horse racecourse at the bend of the river Padma. The area was abandoned for a long period after the British left. Abul Hasnat Muhammad Qamaruzzaman who was a local political leader at that time took initiative to transform the area into a park by Zilla Parishad. The park was inaugurated in 1972 which later handed over to Rajshahi City Corporation in 1996. The zoo inside the park was opened in 1983. Flora and fauna The park has wide varieties of trees. The zoo inside the park has many types of animals including Tiger, deer, crocodile, monkey, snakes. There a many bird species and also fish species in the lake of the park. References Category:Rajshahi Category:Parks in Bangladesh Category:Zoos in Bangladesh Category:1972 establishments in Bangladesh
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Graeme Allan Graeme 'Gubby' Allan (born 23 July 1954) is a former Australian rules footballer who played with Fitzroy and Collingwood in the Victorian Football League (VFL). Allan started his career in the Victorian Football Association where he took the field for Sunshine, and he finished third in the J. J. Liston Trophy count in 1974. He debuted at Fitzroy in 1975 and became a good utility player, often used as a centreman. He was also handy around goals and kicked 25 of them in 1977 and another 26 in 1979. Collingwood acquired his services in 1981 and he participated in that year's Grand Final, on the half back flank. During his time in the VFL he represented the league at interstate football. Since retiring as a player, Allan has continued to be involved in football as an administrator, at Collingwood, Brisbane Lions, and StKilda Football Clubs. His son, Marcus, played briefly at Brisbane, and now plays for the Old Xaverians in the VAFA, along with his youngest son, Patrick "Pat" Allan. References Holmesby, Russell and Main, Jim (2007). The Encyclopedia of AFL Footballers. 7th ed. Melbourne: Bas Publishing. External links Category:1954 births Category:Living people Category:Australian rules footballers from Victoria (Australia) Category:Fitzroy Football Club players Category:Collingwood Football Club players Category:Sunshine Football Club (VFA) players Category:Victorian State of Origin players Category:Collingwood Football Club administrators Category:Brisbane Lions administrators Category:Greater Western Sydney Giants administrators
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Anil Bachoo Anil Kumar Bachoo (born आनिल कुमार बचू on 6 September 1953) is a former Vice Prime Minister and Minister of Public Infrastructure, National Development Unit, Land Transport and Shipping of Mauritius. He has been in office since 13 September 2008 and has already held the office from 2000 to 2005 under the leadership of then MSM/MMM coalition government. From 12 July 2005 to 13 September 2008, he was the Minister of Environment. He is a prominent leader from the Mauritian Labour Party and also a member of parliament for the constituency no 9, Flaq & Bon Acceuil since September 2000. He had previously been MP from 1991 to 1995 until his party alliance MSM/MTD/RMM lost elections to the Mauritian Labour Party alliance . References Category:1953 births Category:Living people Category:Mauritian politicians Category:Vice Prime Ministers of Mauritius
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Tissa Wijesurendra Tissa Wijesurendra (born 3 April 1949 as Sinhala: තිස්ස විජේසුරේන්ද්ර) is an actor in Sri Lankan cinema. One of the most popular actors in 1970s, Wijesurendra has involved in many dramatic roles in a career spanning four decades. Personal life He was born on 3 April 1949. His father was an engineer. He was educated at Nalanda College Colombo. He qualified to become a doctor but became an actor instead. He is married to Savithri Wijesurendra, a retired teacher who formerly worked at Visakha Vidyalaya. The couple has three daughters. Their daughter Surani has been involved with few television serials and has worked as a TV host. Cinema career During school times, he stayed at an uncle's house in Bambalapitiya because it was easy to attend tuition classes. That uncle worked at Ceylon Theaters where he was selected to a film. His maiden cinema appearance came through the 1971 film Kalana Mithuro directed by Kingsley Rajapakse. It was a remake of popular Hollywood film Dosthi. However his first screened film is Nim Walalla in 1970 directed by Ranjith Lal. From the films he acted in, the first box office record was made by me and Malini in the film Suhada Pethuma. Since then, Tissa-Malani era shaped the seventies. After the film Suhada Pethuma, the two stars who were lovers in the films Sadhana, Nila and Ran Kurullo. Also he played the role as lover to many other film actresses like Geetha Kumarasinghe, Nita Fernando, Jenita Samaraweera in films like Sureka, Neela, Niluka, Sudu Paraviyo, Kalana Mithuro and Lasanda. His best contemporaries in the cinema were Vijaya Kumaranatunga, Gamini Fonseka and Tony Ranasinghe. He is the only Sri Lankan actor to continue playing the protagonist in all films he acted since the maiden film. In 1983, he acted in the film Senehasaka Kandulu where he and Sonia Disa played Veddah characters. He expected an award for this character and it premiered in India as well. However, he did not receive the award and temporarily quit from cinema after frustration. In 1993, he played the older role as Sabeetha's father in the film Bambasara Bisaw. Filmography References External links Category:Living people Category:Sinhalese male actors Category:Sri Lankan male film actors Category:Sri Lankan Buddhists Category:Alumni of Nalanda College, Colombo Category:1949 births
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Gumehi Gumehi (, also Romanized as Gūmeh’ī) is a village in Zarrineh Rural District, Karaftu District, Divandarreh County, Kurdistan Province, Iran. At the 2006 census, its population was 360, in 66 families. References Category:Towns and villages in Divandarreh County
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
2012 Prosperita Open The 2012 Prosperita Open was a professional tennis tournament played on clay courts. It was the ninth edition of the tournament which was part of the 2012 ATP Challenger Tour. It took place in Ostrava, Czech Republic between April 30 and May 6, 2012. ATP entrants Seeds 1 Rankings are as of April 23, 2012. Other Entrants The following players received wildcards into the singles main draw: Jakub Lustyk Grzegorz Panfil Adam Pavlásek Jiří Veselý The following players received entry from the qualifying draw: Andrej Martin Marek Michalička Maximilian Neuchrist Grega Žemlja Champions Singles Jonathan Dasnières de Veigy def. Jan Hájek, 7–5, 6–2 Doubles Radu Albot / Teymuraz Gabashvili def. Adam Pavlásek / Jiří Veselý, 7–5, 5–7, [10–8] External links Official Website Prosperita Open Category:Prosperita Open Prosperita Open Prosperita Open Prosperita Open
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Vöcklabrucker SC Vöcklabrucker SC is an Austrian association football club from Vöcklabruck. History After the dissolution of the 1. FC Vöcklabruck and the relegation from the First Division to the regional league. The city has on Hausruck founded a new club, the Vöcklabrucker SC start in the 2nd Klasse Mitte/West. Class mid-west. Current squad www.vbsc.at References External links Official site Category:Football clubs in Austria Category:Association football clubs established in 2009 Category:2009 establishments in Austria
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Ohio State Route 771 State Route 771 (SR 771) is a state highway located entirely in Highland County, Ohio. The north–south route connects SR 138 in Paint Township and SR 28 in Leesburg. Route description The route begins at an unsignalized T-intersection with SR 138 in northern Paint Township. SR 771 heads north-northwest through first lightly wooded areas transitioning to farmland as it enters Fairfield Township. The elevation of the route remains relatively flat except when it dips slightly to cross the Hardin Creek in Fairfield Township. North of the creek, SR 771 enters the village limits of Leesburg and passes the Fairfield High School. While in Leesburg, the road name is South Fairfield Street. As the route enters downtown Leesburg, it passes through residential neighborhoods and light industrial facilities. After crossing the Indiana and Ohio railroad at-grade, SR 771 enters the central business district of the town and ends at a signalized intersection with SR 28. History SR 771 was designated on a SR 138–Leesburg route around 1937 on a gravel road. However, it did not end in downtown Leesburg as it does today; US 62 was routed through the town therefore SR 771 ended at US 62 where County Route 60 intersects the route today. By 1941, the entire route was paved. Between 1969 and 1971, US 62 was rerouted on a bypass around the village which lead to the extension of SR 771 into downtown Leesburg. Since then, no major changes have affected the highway. Major intersections References 771 Category:Transportation in Highland County, Ohio
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Allium tenuiflorum Allium tenuiflorum is a Mediterranean species of wild onion found in Algeria, Morocco, Libya, Italy including Sardinia, and the Balkans. Allium tenuiflorum produces a bulb up to 20 mm long. Scape is up to 40 cm tall. Umbel is lax with uneven pedicels. Flowers are bell-shaped, tepals white with green or purple midveins and white anthers. Ovary at flowering time yellow-green. References tenuiflorum Category:Onions Category:Plants described in 1811
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Bloodsucking Bastards Bloodsucking Bastards is a 2015 American comedy horror film directed by Brian James O'Connell, written by Ryan Mitts and Dr. God, O'Connell's comedy group. The film stars Fran Kranz, Pedro Pascal, Emma Fitzpatrick, and Joey Kern. The film was limited released September 4, 2015, and through video on demand by Scream Factory. Plot At the office where they work together, acting sales manager Evan Sanders talks to his slacker friend and coworker Tim about an upcoming presentation for the Phallucite account. After an awkward moment between Evan and head of HR Amanda in the break room, Tim explains to coworkers Andrew and Mike that Amanda said “I love you” to Evan and Evan’s response was, “no.” Branch president Ted Plunkett passes over Evan for a promotion and instead hires Max Phillips as the new sales manager. Max went to college with Evan and Tim, but Evan had him kicked out after Max slept with his girlfriend. Max moves into Evan’s office and starts hitting on Amanda. In the office alone playing video games after hours, Mike is killed in one of the bathroom stalls. Evan finds Mike’s body in the morning, but it is gone by the time he alerts everyone. Max secretly turns office employee Dave into a vampire. Formerly passive Dave becomes aggressive in demanding that everyone pay what is owed to the office sports betting pool. Anxious to complete his Phallucite presentation, Evan uses Zabeth, who harbors a crush on him, to retrieve files from the basement. Zabeth is attacked and turned into a vampire. Evan convinces Andrew to work late with him. Andrew goes to the basement and is attacked by Zabeth. Meanwhile, Evan searches Max’s office, finding photos of Amanda and personnel files where each employee photograph is marked with an x, circle, or check. Andrew returns upstairs and seemingly drops dead in front of Evan. Evan hides in a supply closet. Determining that Max is behind the murders, Evan calls Amanda with a warning, but Max answers her phone and taunts him. Evan eventually passes out. Ted, Tim, and Max find Evan in the morning. Andrew returns as a vampire and presents the completed Phallucite presentation to the bosses. Ted admonishes Evan for not finishing it himself. However, Max strangely defends Evan to protect Evan’s job. Evan tries convincing Tim that vampires have overtaken the office. Tim reveals several odd events he witnessed that corroborate Evan’s wild claim, including watching Max turn administrative assistant Elaine into a vampire. Amanda still refuses to believe that anything supernatural is going on. Evan worries that Max will turn her into a vampire like the other employees. Evan and Tim go to security guard Frank for help. Frank reveals that he was with Tim when they witnessed the ongoing vampire activity. The three men go to Frank’s car for weapons and discover that Frank’s vehicle was robbed. The janitor reveals he is a vampire by attacking Frank. Frank recovers and stakes the Janitor through his heart, killing him. Evan, Tim, and Frank return to the office and discover that all of the employees are now productive vampires. The men arm themselves with office supplies before confronting Ted, Max, and Andrew in Ted’s office. Amanda is also present. Ted reveals that he knew Max was turning employees into vampires and that he sanctioned it so that employee performance would improve. Evan reveals that the files he found in Max’s office prove that Max planned to kill Ted. Before Ted can act, Max snaps Ted’s neck. Max and Andrew allow the three men to leave Ted’s office with
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Amanda, but they are surrounded by the vampire employees. After a long battle, all of the vampires are killed. Max calls in the legal team, also turned into vampires, as reinforcements. He then compels Amanda back into Ted’s office and leaves the three men to continue fighting. Andrew kills vampire Dave because he never liked him and ends up in a confrontation with Tim during which they debate how their faceoff should end. Frank dies combating the vampires from Legal. Evan returns to Ted’s office where he and Amanda ultimately kill Max. Afterwards, Tim finds Evan and Amanda covered in blood and making out amidst the carnage. The three of them calmly leave the building as day breaks. A cleaning woman arrives at the grisly scene and is attacked and bitten by Andrew. Cast Fran Kranz as Evan Joey Kern as Tim Emma Fitzpatrick as Amanda Marshall Givens as Frank Pedro Pascal as Max Joel Murray as Ted Justin Ware as Andrew Yvette Yates as Zabeth Neil Garguilo as Mike David F. Park as Dave Parvesh Cheena as Jack Sean Cowhig as Janitor Zabeth Russell as Elaine Patricia Rae as Sofia Release Bloodsucking Bastards had its World Premiere as the opening night film of the 2015 Slamdance Film Festival in Park City, Utah. Shortly after it was announced, Scream Factory, had acquired distribution rights to the film. The film went on to premiere at the Sarasota Film Festival on April 11, 2015. and the Texas Frightmare Weekend on May 1, 2015. The film was released on September 4, 2015, in a limited release and through video on demand. Reception As of January 2016, it has received a score of 50% on Metacritic, and 63% on Rotten Tomatoes References External links Interview with Dr. God member David F Park about Bloodsucking Bastards Category:2015 horror films Category:American comedy horror films Category:2015 films Category:2010s comedy horror films Category:American business films Category:American films Category:English-language films Category:Workplace comedies Category:Vampire comedy films
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Alamo Oaks, California Alamo Oaks is an unincorporated community in Contra Costa County, California, United States. It is located in the hills near Alamo in the San Ramon Valley. It lies at an elevation of 515 feet (157 m). The ZIP Code is 94507. The community is inside area code 925. References Category:Unincorporated communities in California Category:Unincorporated communities in Contra Costa County, California
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Race and the War on Drugs The War on Drugs is a term for the actions taken and legislation enacted by the United States government, intended to reduce or eliminate the production, distribution, and use of illicit drugs. The War on Drugs began during the Nixon Administration with the goal of reducing the supply of and demand for illegal drugs, though an ulterior, racial motivation has been proposed. The War on Drugs has led to controversial legislation and policies, including mandatory minimum penalties and stop-and-frisk searches, which have been suggested to be carried out disproportionately against minorities. The effects of the War on Drugs are contentious, with some suggesting that it has created racial disparities in arrests, prosecutions, imprisonment and rehabilitation. Others have criticized the methodology and conclusions of such studies. In addition to enforcement disparities, some claim that the collateral effects of the War on Drugs have established forms of structural violence, especially for minority communities. Motivation for the War on Drugs The War on Drugs was declared by U.S. President Richard M. Nixon during a Special Message to Congress delivered on June 17, 1971, in response to increasing rates of death due to narcotics. During this announcement, Nixon distinguished between fighting the war on two fronts—the supply front and the demand front. To address the "supply" front, Nixon requested funding to train narcotics officers internationally, and proposed various legislation with the intent of disrupting illegal drug manufacturers. The "demand" front referred to enforcement and rehabilitation. To that end, Nixon proposed the creation of the Special Action Office of Drug Abuse Prevention, with the goal of coordinating various agencies in addressing demand for illegal drugs. He also requested an additional $155 million for treatment and rehabilitation programs, and additional funding to increase the size and technological capability of the Bureau of Narcotics and Dangerous Drugs. Alternative motivation A former Nixon aide has suggested that the War on Drugs was racially and politically motivated. However, because he may have been disillusioned with the Nixon administration following the Watergate scandal, the validity of Ehrlichman's claim is disputed. Controversial policies A number of policies introduced during the War on Drugs have been singled out as particularly racially disproportionate. Mandatory minimums The Anti-Drug Abuse Act of 1986 established a 100:1 sentencing disparity for the possession of crack vs. powder cocaine. Whereas possession of 500 grams of powder cocaine triggered a 5-year mandatory minimum sentence, possession of 5 grams of crack cocaine triggered the same mandatory minimum penalty. In addition, the Anti-Drug Abuse Act of 1988 established a 1-year mandatory minimum penalty for simple possession of crack cocaine, making crack cocaine the only controlled substance for which a first possession offense triggered a mandatory minimum penalty. A 1992 study found that, as a result of mandatory minimum sentencing, blacks and Hispanics received more severe sentences than their white counterparts from 1984 through 1990. In 1995, the United States Sentencing Commission delivered a report to Congress concluding that, because 80% of crack offenders were black, the 100:1 disparity disproportionately affected minorities. The Commission recommended that the crack-to-powder sentencing ratio be amended, and that other sentencing guidelines be re-evaluated. These recommendations were rejected by Congress. By contrast, certain authors have pointed out that the Congressional Black Caucus backed the Anti-Drug Abuse Act of 1986, implying that that law could not be racist. In 2010, Congress passed the Fair Sentencing Act, which reduced the sentencing disparity between crack and powder cocaine from 100:1 to 18:1. The mandatory minimum penalty was amended to take effect for possession of crack cocaine in excess of 28 grams. Mandatory minimum penalties
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
have been criticized for failing to apply uniformly to all cases of illegal behavior. Stop-and-frisk searches The Supreme Court ruled in Terry v. Ohio that a "stop-and-frisk" search does not violate the Fourth Amendment, so long as the officer executing the search bears a "reasonable suspicion" that the person being searched has committed, or is about to commit, a crime. As a result, "stop-and-frisk" searches became much more common during the War on Drugs, and were generally conducted in minority communities. "Stop-and-frisk" searches have been criticized for being disproportionately carried out against minorities as a result of racial bias, though empirical literature on this count is inconclusive. Certain authors have found that, even after controlling for location and crime participation rates, African-Americans and Hispanics are stopped more frequently than whites. Others have found that no bias exists, on average, in an officer's decision to stop a citizen, though bias may exist in an officer's decision to frisk a citizen. Stop-and-frisk searches in New York City have declined significantly since 2011. Crime statistics Stops and searches A 2015 report conducted by the Department of Justice found that blacks in Ferguson, Missouri were over twice as likely to be searched during vehicle stops, despite being found in possession of contraband 26% less often than white drivers. A 2016 report conducted by the San Francisco District Attorney's Office concluded that racial disparities exist regarding stops, searches, and arrests by the San Francisco Police Department, and that these disparities were especially salient for the black population. Blacks made up almost 42% of all non-consensual searches after a stop, though they accounted for less than 15 percent of all stops in 2015. Blacks held the lowest search "hit rate", meaning that contraband was least likely to be found during a search. A 2016 Chicago Police Accountability Task Force report found that black and Hispanic drivers were searched by the Chicago Police more than four times more frequently than white drivers, despite white drivers being found with contraband twice as often as black and Hispanic drivers. Arrests A 1995 Bureau of Justice Statistics report found that from 1991 to 1993, 16% of those who sold drugs were black, but 49% of those arrested for doing so were black. A 2006 study concluded that blacks were significantly over-represented among those arrested for drug delivery offenses in Seattle. The same study found that this was a result of law enforcement focusing on crack offenders, on outdoor venues, and dedicating resources to racially heterogeneous neighborhoods. A 2010 study found little difference across races with regards to the rates of adolescent drug dealing. A 2012 study found that African American youth were less likely than white youth to use or sell drugs, but more likely to be arrested for doing so. A 2013 study by the American Civil Liberties Union determined that a black person in the United States was 3.73 times more likely to be arrested for marijuana possession than a white person, even though both races have similar rates of marijuana use. Iowa had the highest racial disparity of the fifty states. Black people in Iowa were arrested for marijuana possession at a rate 8.4 times higher than white people. One factor that may explain the difference in arrest rates between whites and blacks is that blacks are more likely than whites to buy marijuana outdoors, from a stranger, and away from their homes. A 2015 study concluded that minorities have been disproportionately arrested for drug offenses, and that this difference "cannot be explained by differences in drug offending, non-drug offending, or residing in the kinds of neighborhoods
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
likely to have heavy police emphasis on drug offending." Sentencing In 1998, there were wide racial disparities in arrests, prosecutions, sentencing and deaths. African-Americans, who only comprised 13% of regular drug users, made up for 35% of drug arrests, 55% of convictions, and 74% of people sent to prison for drug possession crimes. Nationwide African-Americans were sent to state prisons for drug offenses 13 times more often than white men. Crime statistics show that in 1999 in the United States blacks were far more likely to be targeted by law enforcement for drug crimes, and received much stiffer penalties and sentences than whites. A 2000 study found that the disproportionality of black drug offenders in Pennsylvania prisons was unexplained by higher arrest rates, suggesting the possibility of operative discrimination in sentencing. A 2008 paper stated that drug use rates among Blacks (7.4%) were comparable to those among Whites (7.2%), meaning that, since there are far more White Americans than Black Americans, 72% of illegal drug users in America are white, while only 15% are black. According to Michelle Alexander, the author of The New Jim Crow and a professor of law at Stanford Law School, even though drug trading is done at similar rates all over the U.S., most people arrested for it are colored. Together, African American and Hispanics comprised 58% of all prisoners in 2008, even though African Americans and Hispanics make up approximately one quarter of the US population The majority of prisoners are arrested for drug related crime, and in at least 15 states, 3/4 of them are black or Latino people. A 2012 report by the United States Sentencing Commission found that drug sentences for black men were 13.1 percent longer than drug sentences for white men between 2007 and 2009. Rehabilitation Professor Cathy Schnieder of International Service at American University notes that in 1989, African Americans, representing 12-15 percent of all drug use in the United States, made up 41 percent of all arrests. That is a noted increase from 38 percent in 1988. Whites were 47 percent of those in state-funded treatment centers and made up less than 10 percent of those committed to prison. Percent incarcerated by race and ethnicity Legal history Some have suggested that certain Supreme Court rulings related to the War on Drugs have reinforced racially disproportionate treatment. In United States v. Armstrong (1996), the Supreme Court heard the case of Armstrong, a black man charged with conspiring to possess and distribute more than 50 grams of crack cocaine. Facing the District Court, Armstrong claimed that he was singled out for prosecution due to his race and filed a motion for discovery. The District Court granted the motion, and required the government to provide statistics from the prior 3 year on similar crimes, dismissing Armstrong's case when the government refused. The government appealed the decision, and the U.S. Court of Appeals affirmed the dismissal, holding that defendants in selective-prosecution claims need not demonstrate that the government failed to prosecute similarly situated individuals. The case was then sent to the Supreme Court, which reversed the decision and held that defendants must show that the government failed to prosecute similarly situated individuals. In United States v. Bass (2002), the Supreme Court heard a similar case. John Bass was charged with two counts of homicide, and the government sought the death penalty. Bass filed for dismissal, along with a discovery request alleging that death sentences are racially motivated. When the government refused to comply with the discovery request, the District Court dismissed the death penalty notice. Upon appeal, the U.S. Court of
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Appeals affirmed the dismissal, and the case was sent to the Supreme Court. Reversing the decision, the Supreme Court ruled that a defendant in a selective prosecution case must make a "credible showing" of evidence that the prosecution policy in question was intentionally discriminatory. The court ruled that Bass did not do so because he failed to show that similarly situated individuals of different races were treated differently. Specifically, the Court rejected Bass' use of national statistics, holding that they were not representative of similarly situated individuals. Both cases have been criticized for perpetuating racially motivated legal standards. It has been suggested that the current standard is impossible to meet for selective prosecution claims, because the relevant data may not exist, and if it does, because the prosecution may have sole access to it. Effects of the War on Drugs Negative health effects Felony drug convictions often lead to circumstances that carry negative health-related consequences. Employment opportunities (and associated healthcare benefits), access to public housing and food stamps, and financial support for higher education are all jeopardized, if not eliminated, as a result of such a conviction. In addition, a felony distribution charge often precludes a convict from benefiting from most healthcare programs that receive federal funding. Collateral consequences Some authors have suggested that the collateral consequences of criminal conviction are more serious than the legal penalties. In many cases, statutes do not require that convicts are informed of these consequences. Many felons cannot be employed by the federal government or work in government jobs, as they do not meet the standards to gain security clearance. Felons convicted of distributing or selling drugs may not enlist in the military. Certain states are financially incentivized to exclude criminals from access to public housing. All states receive less federal highway funding if they fail to revoke or suspend driver's licenses of drug-related felons. Collateral consequences, and felon disenfranchisement in particular, have historically been at least partially racially motivated. African-American communities The War on Drugs has incarcerated high numbers of African-Americans. However, the damage has compounded beyond individuals to affect African-American communities as a whole, with some social scientists suggesting the War on Drugs could not be maintained without societal racism and the manipulation of racial stereotypes. African-American children are over-represented in juvenile hall and family court cases, a trend that began during the War on Drugs. From 1985 to 1999, admissions of blacks under the age of 18 increased by 68%. Some authors posit that this over-representation is because minority juveniles commit crime more often, and commit more serious crimes. A compounding factor is often the imprisonment of a father. Boys with imprisoned fathers are significantly less likely to develop the skills necessary for success in early education. In addition, African-American youth often turn to gangs to generate income for their families, oftentimes more effectively than at a minimum wage or entry level job. Still, this occurs even as substance abuse, especially marijuana, has largely declined among high school students. In contrast, many black youths drop out of school, are subsequently tried for drug-related crime, and acquire AIDS at disparate levels. In addition, the high incarceration rate has led to the juvenile justice system and family courts to use race as a negative heuristic in trials, leading to a reinforcing effect: as more African-Americans are incarcerated, the more the heuristic is enforced in the eyes of the courts. This contributes to yet higher imprisonment rates among African-American children. High numbers of African American arrests and charges of possession show that although the majority of drug users in the United States are white,
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
blacks are the largest group being targeted as the root of the problem. Furthermore, a study by Andrew Golub, Bruce Johnson, and Eloise Dunlap affirms the racial divide in drug arrests, notably marijuana arrests, where blacks with no prior arrests (0.9%) or one prior arrest (4.3%) were nearly twice as likely to be sentenced to jail as their white counterparts (0.4% and 2.3%, respectively). Harboring these emotions can lead to a lack of will to contact the police in case of an emergency by members of African-American communities, ultimately leaving many people unprotected. Disproportionate arrests in African-American communities for drug-related offenses has not only spread fear but also perpetuated a deep distrust for government and what some call racist drug enforcement policy. Additionally, a black-white disparity can be seen in probation revocation, where black probationers were revoked at higher rates than white and Hispanic probationers in studies as published under The Urban Institute. Women of color The War on Drugs also plays a negative role in the lives of women of color. The number of black women imprisoned in the United States increased at a rate more than twice that of black men, over 800% from 1986 to 1991. During that same period, the percentage of females incarcerated for drug-related offenses more than doubled. In 1989, black and white women had similar levels of drug use during pregnancy. In spite of this, black women were 10 times as likely as white women to be reported to a child welfare agency for prenatal drug use. In 1997, of women in state prisons for drug-related crimes, forty-four percent were Hispanic, thirty-nine percent were black, and twenty-three percent were white, quite different from the racial make up shown in percentages of the United States as a whole. Statistics in England, Wales, and Canada are similar. Women of color who are implicated in drug crimes are "generally poor, uneducated, and unskilled; have impaired mental and physical health; are victims of physical and sexual abuse and mental cruelty; are single mothers with children; lack familial support; often have no prior convictions; and are convicted for a small quantity of drugs". Additionally, these women typically have an economic attachment to, or fear of, male drug traffickers, creating a power paradigm that sometimes forces their involvement in drug-related crimes. Though there are programs to help them, women of color are usually unable to take advantage of social welfare institutions in America due to regulations. For example, women's access to methadone, which suppresses cravings for drugs such as heroin, is restricted by state clinics that set appointment times for women to receive their treatment. If they miss their appointment, (which is likely: drug-addicted women may not have access to transportation and lead chaotic lives), they are denied medical care critical to their recovery. Additionally, while women of color are offered jobs as a form of government support, these jobs often do not have childcare, rendering the job impractical for mothers, who cannot leave their children at home alone. However, with respect to mandatory minimum sentencing, female offenders receive relief almost 20% more often than male offenders. In addition, female offenders, on average, receive lighter sentences than those who commit similar offenses. See also Methamphetamine and Native Americans Race and crime Race and crime in the United States Race in the United States criminal justice system Racial profiling in the United States United States incarceration rate References Further reading Books The New Jim Crow (2010) by Michelle Alexander www.newjimcrow.com Jill McCorkel (2013). Breaking Women: Gender, Race, and the New Politics of Imprisonment. New York University Press. Journal articles Conference
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
papers News Webb, Gary. "War on drugs has unequal impact on black Americans." (Archive) San Jose Mercury News. August 20, 1996. Category:Drug policy of the United States Category:Race and crime in the United States Category:Race in the United States Category:Social constructionism Category:Social problems in medicine
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Maija Einfelde Maija Einfelde (born 2 January 1939) is a Latvian composer. Biography Maija Einfelde was born in Valmiera, Latvia, and began her music studies with her mother who was a church organist. She continued her education at Alfrēds Kalniņš Music School in Cēsis, at Jāzeps Mediņš Music College in Riga, and then entered the Conservatoire of Latvia in 1966, studying composition with Jānis Ivanovs. After graduating, she has taught music theory and composition at Alfrēds Kalniņš Music School in Cēsis, Emīls Dārziņš Music College and Jāzeps Mediņš Music College. She has one son, writer Jānis Einfelds. Einfelde's works have been performed internationally, including performances by the Brigham Young University Choir, the Vancouver Chamber Choir, the Kansas City Chorale in USA, the Radio Choir of the Netherlands, and at Carnegie Hall, New York City. Honors and awards First Prize, Barlow Endowement for Music Competition, 1997 Grand Music Prize of Latvia, 1997 Award of the Ministry of Culture of the Republic of Latvia, 1999 Works Einfelde composes mainly choral, organ and chamber music. Selected works include: 1982 Sonāte meditācija (Sonata-Meditation) for viola and piano 1985 Second Sonata for violin and piano 1989 Crucifixus for solo organ 1990 Lullaby, for women's choir 1993 String Quartet for two violin, viola and violoncello 1994 Ave Maria for organ 1994 Adagio for violin, cello, and piano 1994 Pirms saules rieta (Before the Sunset) for clarinet, viola and piano 1995 Ave Maria for woman's choir and organ 1996 Pie zemes tālās... (At the Edge of the Earth...), chamber oratorio for mixed choir 1998 Ave Maria for mixed choir and organ 1998 Psalm 15 for mixed choir 1999 Sanctus for solo organ 1999 Prelūdija (Prelude) for oboe and viola 2000 Skerco (Scherzo) for solo violoncello 2002 Noktirne (Nocturne) for harp 2003 Cikls ar Friča Bārdas dzeju (Three Poems by Fricis Bārda) for mixed choir 2006 Divas mīlas dziemsas (Two Love Songs) 12 voices 2002 Symphony Her compositions have been recorded and issued on CD including: No Tevis (From You) // Female Choir Dzintars – 1995, Riga Recording Studio Ave Maria // Female Choir Dzintars – [1996/97] Trīs jūras dziesmas (Three Songs of the Sea) for oboe, French horn, and string orchestra // Rīgas kamermūziķi [Compilation of 20th Century Latvian music] – 1998, BRIZE Gloria for piccolo trumpet and organ / mit Edwart H. Tarr, Irmtraud Krüger, 1998, KREUZ PLUS: MUSIK, BEST. NR. 1627 Pie zemes tālās... (At the Edge of the Earth…) / Latvian Radio Chamber Singers, Conductor Kaspars Putniņš – [Author CD] – 1999 15th Psalm / Latvian Radio Choir, Conductor Sigvards Kļava – 8'49" / Latviešu mūzika Rīgā – 1999 Maija balāde (May Ballad) // Latvian Millennium Classics – 2000, UPE CLASSICS Noktirne (Nocturne) / Divejāda saule tek. Latvian New Music – 2003, LR CD 043 Monologs (Monologue) for cello and piano // Spēlē Māris Villerušs CD 1 – 2003 Dramatiskais dialogs // Negaidīta atklāsme (Unexpected Inspiration) – 2003, Riga Recording Company Sirēnu sala (The Isle of the Sirens) / Vocal ensemble Putni, artistic leader Antra Dreģe // Vokālais ansamblis Putni. Pamošanās – 2003, PUTNI Sonata for Violin and Organ / Jānis Bulavs and Larisa Bulava // Latvian New Music. Latviešu jaunā mūzika – 2004, Jānis Bulavs [recorded in Canada] // Saules lēkts – 2005, LMIC& Latvian Radio Choir, LMIC CD-2005-3 Ave Maria // Latvian String Quartets. Latviešu stīgu kvarteti – 2008 References Category:1939 births Category:Living people Category:People from Valmiera Category:Music educators Category:20th-century classical composers Category:Latvian composers Category:Female classical composers Category:20th-century women musicians Category:Women music educators
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Nexus Q Nexus Q is a digital media player developed by Google. Unveiled at the Google I/O developers' conference on June 27, 2012, the device was expected to be released to the public in the United States shortly thereafter for US$300. The Nexus Q was designed to leverage Google's online media offerings, such as Google Play Music, Google Play Movies & TV, and YouTube, to provide a "shared" experience. Users could stream content from the supported services to a connected television, or speakers connected to an integrated amplifier, using their Android device and the services' respective apps as a remote control for queueing content and controlling playback. The Nexus Q received mixed reviews from critics following its unveiling. While its unique spherical design was praised, the Nexus Q was criticized for its lack of functionality in comparison to similar devices such as Apple TV, including a lack of support for third-party content services, no support for streaming content directly from other devices using the DLNA standard, as well as other software issues that affected the usability of the device. The unclear market positioning of the Nexus Q was also criticized, as it carried a significantly higher price than competing media players with wider capabilities; The New York Times technology columnist David Pogue described the device as being 'wildly overbuilt' for its limited functions. The Nexus Q was given away at no cost to attendees of Google I/O, but the product's consumer launch was indefinitely postponed the following month, purportedly to collect additional feedback. Those who had pre-ordered the Nexus Q following its unveiling received the device at no cost. The Nexus Q was quietly shelved in January 2013, and support for the device in the Google Play apps was phased out beginning in May 2013. Some of the Nexus Q's concepts were repurposed for a more-successful device known as Chromecast, which similarly allows users to wirelessly queue content for playback using functions found in supported apps, but is designed as a smaller HDMI dongle with support for third-party services. Development An early iteration of the Nexus Q was first demoed at Google I/O in 2011 under the name "Project Tungsten"; the device could stream music wirelessly from another Android device to attached speakers. It served as a component of a home automation concept known as "Android@Home", which aimed to provide an Android-based framework for connected devices within a home. Following the launch of the Google Music service in November 2011, a decision was made to develop a hardware device to serve as a tie-in—a project that eventually resulted in the Nexus Q. Google engineering director Joe Britt explained that the device was designed to make music a "social, shared experience", encouraging real-world interaction between its users. He also felt that there had been "a generation of people who’ve grown up with white earbuds", who had thus not experienced the difference of music played on speakers. The Nexus Q was the first hardware product developed entirely in-house by Google, and was manufactured in a U.S.-based factory—which allowed Google engineers to inspect the devices during their production. Hardware and software The Nexus Q takes the form of a sphere with a flat base; Google designer Mike Simonian explained that its form factor was meant to represent a device that pointed towards "the cloud", and "people all around" to reflect its communal nature. The sphere is divided into two halves; the top half can be rotated to adjust the audio volume being output over attached speakers or to other home theater equipment, and tapped to mute. In between the two halves is a
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
ring of 32 LEDs; these lights serve as a music visualizer that animate in time to music, and can be set to one of five different color schemes. The rear of the device contains a power connector, ethernet jack, micro HDMI and optical audio outputs, banana plugs for connecting speakers to the device's built-in 25-watt "stereo-grade" amplifier, and a micro USB connector meant to "connect future accessories and encourage general hack-ability". The Nexus Q includes an OMAP4 processor, 1 GB of RAM, and 16 GB of storage used for caching of streamed content. It also supports near-field communication and Bluetooth for pairing devices and initial setup. The Nexus Q runs a stripped-down version of Android 4.0 "Ice Cream Sandwich", and is controlled solely via supported apps on Android devices running Android 4.1 "Jelly Bean". Google announced plans to support older versions of Android following the device's official launch. Media could be queued to play on the device using a "Play to" button shown within the Google Play Music, Google Play Movies & TV, and YouTube apps. Content is streamed directly from the services by the Nexus Q, with the Android device used like a remote control. For music, multiple users could collaboratively queue songs from Google Play Music onto a playlist. A management app could be used to adjust Nexus Q hardware settings. Nexus Q did not support any third-party media services, nor could media be stored to the device, or streamed to it using the standardized DLNA protocol. Reception Most criticism of the Nexus Q centered on its relatively high price in comparison to contemporary media streaming devices and set-top boxes, such as Apple TV and Roku, especially considering its lack of features when compared to these devices. The New York Times technology columnist David Pogue described the Nexus Q as being a "baffling" device, stating that it was "wildly overbuilt for its incredibly limited functions, and far too expensive", and arguing that it would probably appeal only to people "whose living rooms are dominated by bowling ball collections." Engadget was similarly mixed, arguing that while it was a "sophisticated, beautiful device with such a fine-grained degree of engineering you can't help but respect it", and that its amplifier was capable of producing "very clean sound", the Nexus Q was a "high-price novelty" that lacked support for DLNA, lossless audio, and playback of content from external or internal storage among other features. Discontinuation Nexus Q units were distributed as a gift to attendees of Google I/O 2012, with online pre-orders to the public opening at a price of US$300. On July 31, 2012, Google announced that it would delay the official launch of the Nexus Q in order to address early feedback, and that all customers who pre-ordered the device would receive it for free. By January 2013, the device was no longer listed for sale on the Google Play website, implying that its official release had been cancelled indefinitely. Google began to discontinue software support for the Nexus Q in May 2013, beginning with an update to the Google Play Music app, and a similar update to Google Play Movies & TV in June. The Nexus Q has also been the subject of third-party development and experimentation; XDA-developers users discovered means for side-loading Android applications onto the Nexus Q to expand its functionality. One user demonstrated the ability to use a traditional Android home screen with keyboard and mouse input, as well as the official Netflix app. In December 2013, an unofficial build of Android 4.4 "KitKat" based on CyanogenMod code was also released for the Nexus Q,
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
although it was unstable and lacked reliable Wi-Fi support. The Nexus Q received a de facto successor in July 2013 with the unveiling of Chromecast, a streaming device that similarly allows users to queue the playback of remote content ("cast") via a mobile device. Chromecast is contrasted by its compact HDMI dongle form factor, the availability of an SDK that allows third-party services to integrate with the device, and its considerably lower price in comparison to the Nexus Q. In late 2014, Google and Asus released a second Nexus-branded digital media player known as the Nexus Player, which served as a launch device for the digital media player and smart TV platform Android TV. See also Comparison of set-top boxes Google TV References Further reading Gross, Doug, "Google's new Nexus Q: Made in the U.S.A.", CNN, Thu June 28, 2012 Category:Android (operating system) devices Category:Products introduced in 2012 Category:Streaming media systems Category:Networking hardware Category:Digital media players Category:Google Nexus
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Gold (Jasmine album) Gold is the first studio album by the Japanese singer Jasmine, released in two formats, standard and limited, on July 21, 2010 by Sony Music Associated. The limited edition (a digipak) includes a bonus DVD. Track listing From Natalie.mu Singles Charts Release History References External links Oricon Profile: Regular // Limited Sony Music Profile Regular // Limited Category:2010 debut albums Category:Jasmine (Japanese singer) albums Category:Japanese-language albums Category:Sony Music albums
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
List of NRL Grand finals All but 12 of the National Rugby League's (and its predecessors') 108 seasons had the premiers decided by the outcome of one final game, usually called a grand final. The minor premiership is determined by who finishes first at the end of the season before the finals. The premierships not decided by a grand final were determined by league ladder position. Since 1954 the grand final was officially sanctioned as the only way to determine the premiership, but it had been commonplace for decades before that. 2007 and 2009 premierships were subsequently stripped from Melbourne as a penalty for breaching the salary cap. See also List of NRL Women's Grand finals References Category:NRL Grand Finals Grand finals
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Malta Independent School District Malta Independent School District is a public school district based in the community of Malta, Texas (USA). The district has one school that serves students in grades pre-kindergarten through eight. In 2009, the school district was rated "recognized" by the Texas Education Agency. References External links Malta ISD Category:School districts in Texas Category:School districts in Bowie County, Texas
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Diane Katsiaficas Diane Katsiaficas (El Paso, Texas, 23 November 1947) is an American of Greek heritage visual artist who develops multimedia narrative works, and a professor in the Art Department at the University of Minnesota. She uses a variety of techniques and technologies in her work—from gold leaf to laser cutting. Her work ranges from small journal drawings and paintings to large-scale installations and has been shown throughout the United States and Europe. She has received numerous awards, including a Fulbright Scholarship to Greece and two McKnight Foundation Fellowships, and holds a U.S. patent for an outdoor recreational structure. Educational Background B.A.: Chemistry, Smith College, Northampton, MA, 1968. M.A.T.: Art Education, University of Washington, Seattle, WA, 1974. M.F.A.: Painting, University of Washington, Seattle, WA, 1976. Awards Fulbright Scholar Award (two time recipient) McKnight Research Award, 1996–1999 McKnight Foundation Interdisciplinary Artist Fellowship, 1999 McKnight Foundation Visual Artist Fellowship, 1995 Humanities Institute Fellowship, University of Minnesota, 2002 Exhibitions Her exhibitions include: "Visualizing the Instincts of Migration": American Center, Alexandria, Egypt, 2001. "Migrations": Diana Gallery, Athens, Greece 2002. "Hair Stories": MAEP Gallery, Minneapolis Institute of Art, Minneapolis, Minnesota, 2000. "An Allegory of Olives": Terracotta Gallery, Thessaloniki, Greece, January 2004 - February 2004. "Migrations": Holter Museum of Art, Helena, Montana, January 2005 - April 2005. References Selected bibliography Gleason, Norma Catherine. "Exhibitions. Summer Sets." Artweek vol. 13 nº. 29. September 11, 1982. p. 8, illus. Glowen, Ron. "Exhibitions: The Will to Order." Artweek vol. 11 nº 34. October 18, 1980 p. 16, illus. Guenther, Bruce. 50 Northwest Artists: A Critical Selection of Painters and Sculptors Working in the Pacific Northwest. San Francisco: Chronicle Books, 1983. Kangas, Matthew. "Exhibitions Down on the Farm: Environmental Sculpture Invitational" Artweek vol. 12 nº 29. September 12, 1981 p. 3, illus. Toale, Bernard. The Art of Papermaking. Worcester, Massachusetts: Davis, 1983. External links Diane Katsiaficas website Old Website- works until 2004 "An Interview with Diane Katsiaficas", by Mary Hark. Hand Papermaking, Summer 1992, pp. 2–9. In the margins between virtue and vice - Women in the early religions of the Easterna Mediterranean, by Diane Katsiaficas Watson-Jones, Virginia Contemporary American Women Sculptors. Oryx Press. Phoenix, 1986, p. 324 Category:University of Minnesota faculty Category:Living people Category:American artists Category:Smith College alumni Category:University of Washington College of Education alumni Category:Fulbright Scholars Category:Year of birth missing (living people) Category:University of Washington School of Art + Art History + Design alumni
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Kirby (character) () is the titular protagonist of the Kirby series of video games owned by Nintendo and HAL Laboratory. As one of Nintendo's most famous and familiar icons, Kirby's round appearance and ability to copy his foes' powers has made him a well-known figure in video games, consistently ranked as one of the most iconic video game characters. He first appeared in 1992 in Kirby's Dream Land for the Game Boy. Originally a placeholder, created by Masahiro Sakurai, at the age of 19, for the game's early development, he has since then starred in over 20 games, ranging from action platformers to puzzle, racing, and even pinball, and has been featured as a playable fighter in all Super Smash Bros. games. He has also starred in his own anime and manga series. His most recent appearance is in Kirby Star Allies, for the Nintendo Switch. Since 1999, he has been voiced by Makiko Ohmoto. Kirby is well known for his ability to inhale objects and creatures to gain their abilities, as well as his ability to float by expanding his body. He uses these abilities to rescue various lands, including his homeworld of Planet Popstar, from evil forces and antagonists, such as Dark Matter or Nightmare. On these adventures he often crosses paths with his rivals, the gluttonous King Dedede and the mysterious Meta Knight, though Meta Knight can be seen as a hero, depending on the game. Kirby has been described as one of the most legendary video game characters of all time by PC World. In virtually all his appearances, Kirby is depicted as cheerful, innocent and food-loving but becomes fearless, bold and clever in the face of danger. Concept and creation Kirby was created by video game designer Masahiro Sakurai as the player character of the 1992 game Kirby's Dream Land. The character's design was intended to serve as placeholder graphics for the game's original protagonist in early development and thus was given a simplistic ball-like appearance. Sakurai switched to the placeholder design for the final character design when he believed that it served the character better. The character was known as during development until the name "Kirby" was chosen from a draft list of potential names. Shigeru Miyamoto stated that "Kirby" was chosen in honor of American lawyer John Kirby, who defended Nintendo in the Universal City Studios, Inc. v. Nintendo Co., Ltd. case in 1984 and that the guttural-sounding name contrasted amusingly with the character's cute appearance. Kirby appears white in Kirby's Dream Land due to the grayscale palette of the Game Boy system. Sakurai intended the character to be pink, though Miyamoto originally wanted the character to be yellow. Kirby does appear white in the game's North American promotional materials and artwork to reflect the character's in-game appearance. In North America, Kirby consistently appears in artwork with a more striking, bolder expression than in Japan, where his expression is more relaxed. Nintendo has said that the more battle-ready appearance appeals to a wider audience in North America. Characteristics Physical appearance Kirby has a pink, spherical, body with small stubby arms and large red feet. His eyes are a distinctive oval shape and are white (eye shine) at the top, black in the center and dark blue on the bottom (all black in early games) with rosy cheek-blushes near his eyes. His body is soft and flexible, allowing him to stretch or flatten and adopt different shapes, open his mouth really wide to inhale foes, or inflate himself with air and fly. According to the Super Smash Bros. manual he is 8 inches (or
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
20.32 centimeters) tall, however the games themselves are inconsistent with Kirby's height. Kirby's appearance has changed subtly over the years, becoming more rounded and defined, mainly in his face and larger blue eyes. The new design has been used in all subsequent games. Personality Kirby hails from the distant Planet Popstar, where he lives in a dome-shaped house in the country of Dream Land. He has a positive attitude, and frequently helps to save Dream Land through the use of his unique powers. His age is never directly stated, although he is referred to as a "little boy" in the English manual for Kirby's Dream Land, described as a baby in the anime, and was referred to as being a "jolly fellow" in Kirby Super Star. Kirby is cheerful and innocent. He is often depicted with a voracious appetite. His other hobbies include singing, although he is tone-deaf. The same personality and nature is presented in the Super Smash Bros. games. In Super Smash Bros. Melee, his trophy profile states that he is an "extremely skilled technician". Kirby plays a prominent role in the Subspace Emissary plot as a protagonist, the story mode of Super Smash Bros. Brawl. Although he appears cute and innocent, many commercials and ads have contrasted this with his extreme fighting skills when he takes on the abilities of an enemy. Kirby is a character of few words and rarely speaks in-game. His dialogue is largely limited to appearances in some games' instruction manuals, and brief comments such as "Hi!" or "Poyo!" in games including Super Smash Bros. and Kirby 64: The Crystal Shards. Despite his lack of dialogue, Kirby narrates the functions of certain Copy Abilities on the pause menu in Kirby: Nightmare in Dream Land, Kirby & the Amazing Mirror, and Kirby: Squeak Squad. Kirby gives short descriptions of various menu options when they are highlighted. The concept of Kirby being mostly mute is supported by Masahiro Sakurai's insistence that Kirby does not speak in the anime series, although he does occasionally say the names of his special attacks and certain foods and sometimes repeats other words that he hears. In the Super Smash Bros. series, when Kirby copies a foe's ability, Kirby says the name of some attacks, such as Captain Falcon's "Falcon Punch", Ness' PK Flash, Lucas' PK Freeze, Ryu's Hadoken/Shakunetsu Hadoken, "Jump/Speed/Shield/Buster/Smash" when using Shulk's Monado Arts and Terry Bogard's Power Wave. There are some notable exceptions where Kirby actually does speak in-game, such as Kirby's Avalanche, Kirby no Kirakira Kizzu and Kirby's Epic Yarn. Despite his loveable nature, he sometimes acts in an impulsive or naive way, such as when he accuses King Dedede of interfering with the Fountain of Dreams in Kirby's Adventure, and then leaves to retrieve the pieces of the shattered Star Rod without receiving a full explanation. He also attacked Meta Knight in Kirby Squeak Squad for ownership of a chest that he believed contained a slice of his stolen strawberry shortcake. In reality, the Star Rod was going to be taken by Nightmare, who would corrupt the Fountain of Dreams and turn every dream into a Nightmare, and the chest that Kirby wanted contained Dark Nebula, a powerful being with evil intentions. King Dedede and Meta Knight were aware of these problems, and were trying to protect the world. Abilities Kirby's main ability is inhaling enemies and objects and spitting them out with incredible force. Both his mouth and body expand to allow him to inhale things much larger than himself. However, there is a limit to what he can inhale; excessively large or heavy
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
foes like bosses can resist Kirby's inhale. Against these opponents Kirby must find smaller objects to use as ammo or intercept his opponent's attacks and send them back at them. Kirby is able to float like a balloon by inhaling a gulp of air to make himself more buoyant, flying slowly by flapping his arms. However, his only method of attack in this state is to spit the gulp of air, sending him back towards the ground. While in most games he can float indefinitely, he gets tired in some games such as Kirby 64: The Crystal Shards, Team Kirby Clash Deluxe, Kirby Fighters Deluxe and Super Smash Bros.. Kirby's main mode of transportation is a star shaped, flying vehicle called a Warp Star. By clinging to its side or riding it like a surfboard Kirby is able to move quickly through the air and travel quickly between areas. From the second game onwards, by swallowing foes, Kirby can copy their primary attribute, called a Copy Ability. Giving Kirby new abilities such as gaining fire abilities from inhaling a fiery creature or the ability to shoot needles from his body by eating a spiky creature. Kirby can only have one Copy Ability at a time and must discard his current one if he wishes to use another. Additionally, if he takes too much damage his Copy Ability will be knocked out of him; though if he's quick he can inhale it again. Kirby Super Star greatly expanded on the concept of Copy Abilities from Kirby's Adventure. Originally Kirby just obtained one technique from a Copy Ability which replaced his Inhale but Kirby Super Star introduced multiple fighting techniques for each ability with their own controls. Additionally while Kirby's Adventure didn't have Kirby's appearance drastically change when he had a Copy Ability, aside from his color changing to orange or snow white, or wielding a weapon for abilities like Sword or Hammer. Kirby Super Star gave each Copy Ability its own costume, such as a fiery crown for Fire Kirby or a floppy green cap that resembles Link's for Sword Kirby. In Kirby Super Star, Kirby could sacrifice his current Copy Ability to create a 'Helper', an allied version of the enemy Kirby typically gets that ability from. The new Helper could be controlled by the game or a second player. This feature has not yet been included in later Kirby games, although it was included in Kirby Super Star's remake, Kirby Super Star Ultra and was intended to return in the cancelled Gamecube Kirby game which would have had Kirby able to make up to four helpers at once. In the game Kirby Star Allies, the game’s main mechanic is similar to this, but with now up to 3 helpers. Other games have developed his powers even further, often with new ways to utilize his Copy Abilities. As a whole, however, his basic moves have remained the same. In Kirby 64 Kirby can combine two Copy Abilities to create new ones. This returns in Kirby: Squeak Squad in a limited form, but only with Bomb and Sword which are both able to combine with Ice, Fire, and Spark. In Kirby Star Allies, this returns again with more combinations available to the player. In Kirby's Return to Dream Land, some Copy Abilities received enhanced versions called Super Abilities. Stronger than normal Copy Abilities, Super Abilities are capable of destroying scenery and interacting with the environment but unlike normal Copy Abilities their use has a time limit. Likewise, in Kirby: Triple Deluxe, Kirby can eat a Miracle Fruit to become Hypernova Kirby,
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
giving his basic inhale ability a temporary boost that allows him to inhale huge objects. In addition to his natural and Copy Abilities Kirby makes use of a variety of power-ups and items. In Kirby's Dream Land, Kirby's first appearance, Kirby did not have Copy Abilities. He collected power ups, such as a mint leaf, which enabled him to shoot out endless air pellets for a limited amount of time, and super spicy curry, which let Kirby shoot out fireballs from his mouth until the item's effect wore off or he lost a life. Kirby uses food or other special items like Maxim Tomatoes or Invincibility Lollipops to heal his health or gain temporary invincibility in the case of the lollipop. Kirby Super Star, Kirby Super Star Ultra, Kirby & the Amazing Mirror and Kirby's Return to Dream Land contain a feature where Kirby can share his food with the second player. In Super Smash Bros. Brawl, Kirby's Final Smash is his Cook ability from Kirby Super Star, which cook opponents, items and more. In Super Smash Bros. for Nintendo 3DS and Wii U and Super Smash Bros. Ultimate, Kirby's Final Smash is an Ultra Sword attack. In Kirby's Epic Yarn, at the start of the game, Kirby inhales a yarn-based Maxim Tomato (referred to by Yin-Yarn as a Metamato) transforming him into yarn, granting Kirby the ability to transform into a car, a surfboarding penguin, a UFO, and many other forms. Also, since Kirby cannot suck up enemies in this game (since air just goes through him), he uses a yarn whip to grab enemies and turn them into yarn balls, which can be thrown at other foes, attached to buttons, move zippers and more. Species Although never explored in great detail in the series, Kirby is not the only member of his kind. Similar looking characters are seen in the ending of Kirby's Dream Land, and its remake Spring Breeze in Kirby Super Star. Meta Knight and Galacta Knight, both of whom bear a close resemblance to Kirby without their masks are often hinted to be of the same species as Kirby. In general, members of the species all look similar, but sometimes have different colored skin, eyes, and feet. No official term exists for them other than Kirby's species. They are commonly referred to as 'Kirbys', just as Yoshi in the Mario series is the name of both the species and character. As a name, however, 'Kirby' refers to only one character. In the English manual of Kirby & the Amazing Mirror, the term 'Kirbys' is used at points, to illustrate the feature of having four differently colored versions of Kirby active in the game simultaneously, although in this situation the four "Kirbys" are four pieces of Kirby rather than four individual Kirby-like beings. Another common term is 'Dream Landers', a term used in the instruction manual for Kirby's Adventure. This term, however, also refers to anyone who lives in Dream Land, including characters like King Dedede. The biography of Kirby in Super Smash Bros. Brawl says Kirby is a citizen of Dream Land. Sometimes the term 'puffballs' is used referring to the species' round shape. In Kirby: Nightmare in Dream Land, Kirby's Return to Dream Land and Kirby & the Amazing Mirror, different colored Kirbys appear as other players in multi-player games. In the case of the latter, they are the result of Kirby being divided into four by Dark Meta Knight. In the anime, Kirby has also been called a Star Warrior, along with Meta Knight and other characters that have appeared in the Kirby
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
animated series. It is said that Kirby is just a young Star Warrior or an underdeveloped one. In the first episode of the animated series, Kirby's space ship prematurely crashed on Pop Star. The trip was supposed to take him several hundred years and in that time he was supposed to be dormant; however, his ship drifted into a space rift causing it to warp to Dream Land and making his trip shorter, not allowing him to develop into the Star Warrior he was going to become. Kirby was also called a Warpstar Knight or a Warpstar Warrior, but this was only occasionally seen in select advertisements. In Kirby's Dream Land 3, a mysterious enemy appears called Batamon, notable for its resemblance to Kirby aside from a more flattened shape and different face. These creatures are always seen walking past walls or ceilings beyond Kirby's reach, with the exception of a single stage in Cloudy Park, where Kirby can continue past the exit and come into contact with some. Appearances Main series Kirby's primary media form is video games, as he is the star and protagonist of the Nintendo owned series named after him. Most Kirby games are platform games in which Kirby fights enemies, solves puzzles, and challenges bosses. Each game typically adds a new twist or change to Kirby's abilities that the levels of the game emphasize, such as combining power ups in Kirby 64 or Super Abilities in Return to Dream Land. Most games involve Kirby saving a world or even universe from a dark, evil force of some kind. Typical Kirby games have a powerful being, such as Dark Matter or Nightmare, controlling or corrupting a secondary villain. The series has 26 games currently, with the first being Kirby's Dream Land in 1992 and the latest being Kirby Star Allies in 2018. Although most are platform games, some spin offs have included pinball, racing, and puzzle games. Super Smash Bros. series Kirby appears in the Super Smash Bros. series of games as a playable character, which is also created by Masahiro Sakurai. He is the sole Kirby series representative in Super Smash Bros. and Super Smash Bros. Melee, and has been one of three (the others being Meta Knight and King Dedede) as of Super Smash Bros. Brawl. Kirby reprises his role as a playable character in Super Smash Bros. for Nintendo 3DS and Wii U and Super Smash Bros. Ultimate, the latter of which features him playing a starring role in the "World of Light" story mode. Anime Kirby appears in his own anime titled Kirby: Right Back at Ya!, (Japanese: 星のカービィ Hoshi no Kaabii) in Japan. The series was licensed in North America by 4Kids Entertainment and the Canadian company Nelvana, and produced by Warpstar Inc., a company formed between a joint investment between Nintendo and HAL Laboratory. It aired on 4Kids TV, via Chubu-Nippon Broadcasting (since 2001 which has done 100 episodes). The anime series is set in its own universe independent of the games and offers its own take on the setting and characters, however, Kirby's creator, Masahiro Sakurai, was greatly involved in its creation, so it did not stray far from his vision of how Kirby should be. The anime contains darker themes such as war, and death while still maintaining the cheerful and surreal themes the games are known for. In the anime Kirby's origin and backstory are explored much more deeply than in the games. Kirby is a legendary Star Warrior who, according to legend, is fated to save Pop Star from destruction, but he was awoken from his
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
slumber 200 years too early by mistake and as such has many baby or childlike qualities. Because of his age, he is still getting the hang of his powers and depends even more on the help of his friends to pull through tough situations. It is also implied that Kirby was originally a creation of the ancient evil entity called Nightmare but refused to follow his evil orders and was discarded into the depths of space. Kirby's Warpstar is also expanded upon, as in the games it is a mode of transportation, but here the Warpstar is also the source of his power but as he's too young to keep it safe for himself the character Tiff takes it upon herself to keep the Warpstar safe. Though Sakurai instructed that Kirby not speak in full sentences (citing the Peanuts character Snoopy) he did allow him to use a small vocabulary including a catch phrase "Poyo" and the names of various attacks and characters. Prior to this Kirby was also in a short educational video that was released exclusively in Japan in 1994, designed to teach Kanji to young children. The feature was not animated, but contained illustrations and was bundled with a similar video featuring Mario and Wario. Manga and comics Kirby also stars in several manga series, most of which have only been released in Japanese. The longest running series is titled 星のカービィ デデデでプププなものがたり (lit. Kirby of the Stars: The Story of Dedede In Pupupu) running up to 25 books, and was written by Hirokazu Hikawa (ひかわ 博一 Hikawa Hirokazu). While Viz Media originally had plans to release the manga in English on September 2010. This release date was pushed back to October 4, 2011 before being canceled completely on May 16, 2011. Other Kirby manga are typically one-shot comedy 4koma gag-manga based on the games, and have multiple artists. The German Club Nintendo magazine often featured comics starring various Nintendo characters including Kirby. In his comics Kirby is reimagined as a detective and King Dedede as his sidekick. These comics were promotions for Kirby games that were released in Germany at the time including Kirby's Dream Land 2, Kirby's Avalanche, and Kirby's Dream Course. Reception Since he first appeared in Kirby's Dream Land, Kirby has received positive reception. Nintendo Power listed Kirby as their 19th favourite hero, commenting that he doesn't get the respect that he deserves. Kirby ranked second on GameDaily's Top 10 Super Smash Bros. characters list. They also included him in their "Pretty in Pink Video Game Characters" article. GamesRadar listed Kirby as one of the most lovable blobs, calling him one of the cutest things to appear in a Nintendo game, yet also describing the way he defeats his enemies as "horrific." He ranked sixth on IGN's top 10 veteran Super Smash Bros. characters, described as the "pinkest badass ever made." An issue of the webcomic VG Cats depicts Kirby in 100 forms from both video games and other forms of media, such as characters from The Legend of Zelda, Metal Gear and Gundam. In 2011, Cheat Code Central ranked Ninja Kirby as the ninth top ninja in video games. UGO Networks listed Kirby on their list of "The Cutest Video Game Characters" stating "It's easy to get on board with someone who will eat anything." In 2009, GameSpot chose him as one of the characters to compete in their poll for the title of "All Time Greatest Game Hero". In a 2010 Famitsu poll, Kirby was voted by readers as the 12th most popular video game character. The 2011 Guinness World Records Gamer's
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Edition lists Kirby as the 18th most popular video game character. In 2012, GamesRadar ranked him as the 40th "most memorable, influential, and badass" protagonist in games, saying "[i]f you don’t think a pink blob can be tough, then you clearly haven’t seen Kirby in action". See also List of Kirby characters References External links Kirby's Official U.S. homepage Kirby on Play.Nintendo.com The Evolution of Kirby Category:Kirby characters Category:Nintendo protagonists Category:Child characters in video games Category:Child superheroes Category:Extraterrestrial characters in video games Category:Extraterrestrial superheroes Category:Fictional characters who use magic Category:Male characters in video games Category:Shapeshifter characters in video games Category:Super Smash Bros. fighters Category:Video game characters introduced in 1992 Category:Video game characters who use magic Category:Video game mascots Category:Video game characters in anime and manga
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Tikmah Ajam Tikmah Ajam (, also Romanized as Tīkmah ‘Ajam) is a village in Qaleh Darrehsi Rural District, in the Central District of Maku County, West Azerbaijan Province, Iran. At the 2006 census, its population was 130, in 30 families. References Category:Populated places in Maku County
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Something to Live For Something to Live For may refer to: "Something to Live For" (song), a 1939 jazz standard by Duke Ellington and Billy Strayhorn Something to Live For (film), a 1952 film directed by George Stevens Something to Live for: The Alison Gertz Story, a 1992 television movie starring Molly Ringwald "Something to Live For", a 2006 episode from the first season of My Name Is Earl Something to Live For (album), a 2007 album by John O'Callaghan Something to Live For: The Music of Billy Strayhorn, a 1987 album by Art Farmer Something to Live For: A Billy Strayhorn Songbook, a 1998 album by John Hicks
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Emperor Tenmu was the 40th emperor of Japan, according to the traditional order of succession. Tenmu's reign lasted from 673 until his death in 686. Traditional narrative Tenmu was the youngest son of Emperor Jomei and Empress Kōgyoku, and the younger brother of the Emperor Tenji. His name at birth was Prince Ōama (大海人皇子:Ōama no ōji). He was succeeded by Empress Jitō, who was both his niece and his wife. During the reign of his elder brother, Emperor Tenji, Tenmu was forced to marry several of Tenji's daughters because Tenji thought those marriages would help to strengthen political ties between the two brothers. The nieces he married included Princess Unonosarara, today known as Empress Jitō, and Princess Ōta. Tenmu also had other consorts whose fathers were influential courtiers. Tenmu had many children, including his crown prince Kusakabe by Princess Unonosarara; Princess Tōchi; Prince Ōtsu and Princess Ōku by Princess Ōta (whose father also was Tenji); and Prince Toneri, the editor of the Nihon Shoki and father of Emperor Junnin. Through Prince Kusakabe, Tenmu had two emperors and two empresses among his descendants. Empress Kōken was the last of these imperial rulers from his lineage. Events of Tenmu's life Emperor Tenmu is the first monarch of Japan, to whom the title Tennō (Emperor of Japan) was assigned contemporaneously—not only by later generations. The only document on his life was Nihon Shoki. However, it was edited by his son, Prince Toneri, and the work was written during the reigns of his wife and children, causing one to suspect its accuracy and impartiality. He is also mentioned briefly in the preface to the Kojiki, being hailed as the emperor to have commissioned them. Tenmu's father died while he was young, and he grew up mainly under the guidance of Empress Saimei. He was not expected to gain the throne, because his brother Tenji was the crown prince, being the older son of their mother, the reigning empress. During the Tenji period, Tenmu was appointed his crown prince. This was because Tenji had no appropriate heir among his sons at that time, as none of their mothers was of a rank high enough to give the necessary political support. Tenji was suspicious that Tenmu might be so ambitious as to attempt to take the throne, and felt the necessity to strengthen his position through politically advantageous marriages. Tenji was particularly active in improving the military institutions which had been established during the Taika reforms. In his old age, Tenji had a son, Prince Ōtomo, by a low-ranking consort. Since Ōtomo had weak political support from his maternal relatives, the general wisdom of the time held that it was not a good idea for him to ascend to the throne, yet Tenji was obsessed with the idea. In 671 Tenmu felt himself to be in danger and volunteered to resign the office of crown prince to become a monk. He moved to the mountains in Yoshino, Yamato Province (now Yoshino, Nara), officially for reasons of seclusion. He took with him his sons and one of his wives, Princess Unonosarara, a daughter of Tenji. However, he left all his other consorts at the capital, Omikyō in Ōmi Province (today in Ōtsu). A year later, (in 672) Tenji died and Prince Ōtomo ascended to the throne as Emperor Kōbun. Tenmu assembled an army and marched from Yoshino to the east, to attack the capital of Omikyō in a counterclockwise movement. They marched through Yamato, Iga and Mino Provinces to threaten Omikyō in the adjacent province. The army of Tenmu and the army of the young Emperor Kōbun
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
fought in the northwestern part of Mino (nowadays Sekigahara, Gifu), an incident known as the Jinshin War. Tenmu's army won and Kōbun committed suicide. Post-Meiji chronology In the 10th year of Tenji, in the 11th month (671): Emperor Tenji, in the 10th year of his reign (天智天皇十年), designated his son as his heir; and modern scholars construe this as meaning that the son would have received the succession (senso) after his father's death. Shortly thereafter, Emperor Kōbun is said to have acceded to the throne (sokui). If this understanding were valid, then it would follow: In the 1st year of Kōbun (672): Emperor Kōbun, in the 1st year of his reign (弘文天皇一年), died; and his uncle Ōaomi-shinnō received the succession (senso) after the death of his nephew. Shortly thereafter, Emperor Tenmu could be said to have acceded to the throne (sokui). Pre-Meiji chronology Prior to the 19th century, Otomo was understood to have been a mere interloper, a pretender, an anomaly; and therefore, if that commonly accepted understanding were to have been valid, then it would have followed: In the 10th year of Tenji, in the 11th month (671): Emperor Tenji, in the 10th year of his reign (天智天皇十年), died; and despite any military confrontations which ensued, the brother of the dead sovereign would have received the succession (senso); and after a time, it would have been understood that Emperor Tenmu rightfully acceded to the throne (sokui). As might be expected, Emperor Tenmu was no less active than former-Emperor Tenji in improving the Taika military institutions. Tenmu's reign brought many changes, such as: (1) a centralized war department was organized; (2) the defenses of the Inner Country near the Capital were strengthened; (3) forts and castles were built near Capital and in the western parts of Honshū—and in Kyushu; (4) troops were reviewed; and all provincial governors were ordered to complete the collection of arms and to study tactics. In 673 Tenmu moved the capital back to Yamato on the Kiymihara plain, naming his new capital Asuka. The Man'yōshū includes a poem written after the Jinshin War ended: At Asuka, Emperor Tenmu was enthroned. He elevated Unonosarara to be his empress. Events of his reign include: 674 (Tenmu 2): Ambassadors of Tane no kuni were received in the Japanese court. 680 (Tenmu 8): Yakushi-ji was founded in the Hakuhō period. Tenmu reigned from this capital until his death in 686. His wife, Empress Jito became the emperor until their son became the 42nd Emperor. The actual site of his grave is known. This emperor is traditionally venerated at a memorial shrine (misasagi) in Nara Prefecture. The Imperial Household Agency designates this location as Tenmu's mausoleum. It is formally named Hinokuma no Ōuchi no misasagi. Buddhism In 675 Emperor Tenmu banned the consumption of domesticated animal meat (horse, cattle, dogs, monkeys, birds), from April 1st to September 30th each year, due to the influence of Buddhism. Wild game meat was exempt. Politics In the Nihon Shoki, Tenmu is described as a great innovator, but the neutrality of this description is doubtful, since the work was written under the control of his descendants. It seems clear, however, that Tenmu strengthened the power of the emperor and appointed his sons to the highest offices of his government, reducing the traditional influence of powerful clans such as the Ōtomo and Soga clans. He renewed the system of kabane, the hereditary titles of duty and rank, but with alterations, including the abolition of some titles. Omi and Muraji, the highest kabane in the earlier period, were reduced in value in the new hierarchy, which consisted
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
of eight kinds of kabane. Each clan received a new kabane according to its closeness to the imperial bloodline and its loyalty to Tenmu. Tenmu attempted to keep a balance of power among his sons. Once he traveled to Yoshino together with his sons, and there had them swear to cooperate and not to make war on each other. This turned out to be ineffective: one of his sons, Prince Ōtsu, was later executed for treason after the death of Tenmu. Tenmu's foreign policy favored the Korean kingdom Silla, which took over the entire Korean peninsula in 676. After the unification of Korea by Silla, Tenmu decided to break diplomatic relations with the Tang dynasty of China, evidently in order to keep on good terms with Silla. Tenmu used religious structures to increase the authority of the imperial throne. During his reign there was increased emphasis on the tie between the imperial household and Ise Grand Shrine (dedicated to the ancestor goddess of the emperors, Amaterasu) by sending his daughter Princess Ōku as the newly established Saiō of the shrine, and several festivals were financed from the national budget. He also showed favor to Buddhism, and built several large temples and monasteries. It is said that Tenmu asked that each household was encouraged to build an altar with a dais where a Buddha-image and a sutra could be placed so that family worshiping could be held, thus inventing the butsudan. On the other hand, all Buddhist priests, monks and nuns were controlled by the state, and no one was allowed to become a monk without the state's permission. This was aimed at preventing cults and stopping farmers from turning into priests. Kugyō Kugyō () is a collective term for the very few most powerful men attached to the court of the Emperor of Japan in pre-Meiji eras. In general, this elite group included only three to four men at a time. These were hereditary courtiers whose experience and background would have brought them to the pinnacle of a life's career. During Tenmu's reign, this apex of the Daijō-kan included: Sadaijin, Soga no Akae no Omi Udaijin, Nakatomi no Kane no Muraji Naidaijin Era of Tenmu's reign The years of Tenmu's reign were marked by only one era name or nengō, which was proclaimed in the final months of the emperor's life; and Shuchō ended with Tenmu's death. Shuchō (686) Non-nengō period The early years of Tenmu's reign are not linked by scholars to any era or nengō. The Taika era innovation of naming time periods – nengō – was discontinued during these years, but it was reestablished briefly in 686. The use of nengō languished yet again after Tenmu's death until Emperor Monmu reasserted an imperial right by proclaiming the commencement of Taihō in 701. See Tenmu period (673–686). In this context, Brown and Ishida's translation of Gukanshō offers an explanation about the years of Empress Jitō's reign which muddies a sense of easy clarity in the pre-Taihō time-frame: "The eras that fell in this reign were: (1) the remaining seven years of Shuchō [(686+7=692?)]; and (2) Taika, which was four years long [695–698]. (The first year of this era was kinoto-hitsuji [695].) ... In the third year of the Taika era [697], Empress Jitō yielded the throne to the Crown Prince." Wives and Children Empress: Princess Uno-no-sarara (鸕野讃良皇女) later Empress Jitō, Emperor Tenji’s daughter Second Son: Prince Kusakabe (草壁皇子), Father of Emperor Monmu and Empress Genshō Hi: Princess Ōta (大田皇女), Emperor Tenji’s daughter Second daughter: Princess Ōku (大伯皇女), Saiō in Ise Shrine (673–686) Third Son: Prince Ōtsu (大津皇子) Hi:
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Princess Ōe (大江皇女), Emperor Tenji’s daughter Seventh Son: Prince Naga (長皇子) Ninth Son: Prince Yuge (弓削皇子) Hi: Princess Niitabe (新田部皇女), Emperor Tenji’s daughter Sixth Son: Prince Toneri (舎人皇子), Father of Emperor Junnin Bunin: Fujiwara no Hikami-no-iratsume (藤原氷上娘, d.682), Fujiwara no Kamatari’s daughter Princess Tajima (但馬皇女), married to Prince Takechi Bunin: Fujiwara no Ioe-no-iratsume (藤原五百重娘), Fujiwara no Kamatari’s daughter Tenth Son: Prince Niitabe (新田部皇子, d.735) Bunin: Soga no Ōnu-no-iratsume (蘇我大蕤娘), Soga no Akae’s daughter Fifth Son: Prince Hozumi (穂積皇子) Princess Ki (紀皇女) Princess Takata (田形皇女), Saiō in Ise Shrine (706–707), later married to Prince Mutobe Court lady: Princess Nukata (額田王), Prince Kagami's daughter First Daughter: Princess Tōchi (十市皇女), married to Emperor Kōbun Court lady: Munakata no Amako-no-iratsume (胸形尼子娘), unakata-no-Kimi Tokuzen's daughter First Son: Prince Takechi (高市皇子) Court lady: Shishihito no Kajihime-no-iratsume (宍人梶媛娘), Shishihito-no-Omi Ōmaro's daughter Fourth Son: Prince Osakabe (刑部/忍壁皇子) Princess Hatsusebe (泊瀬部皇女), married to Prince Kawashima (son of Emperor Tenji) Princess Taki (託基皇女/多紀皇女) (?–751), Saiō in Ise Shrine (698–before 701), later married to Prince Shiki (son of Emperor Tenji) Prince Shiki (磯城皇子) Ancestry See also Emperor of Japan List of Emperors of Japan Imperial cult References Further reading Asakawa, Kan'ichi. (1903). The Early Institutional Life of Japan. Tokyo: Shueisha. ; see online, multi-formatted, full-text book at openlibrary.org Aston, William. (1896). Nihongi: Chronicles of Japan from the Earliest Times to A.D. 697. London: Kegan Paul, Trench, Trubner. Brown, Delmer M. and Ichirō Ishida, eds. (1979). Gukanshō: The Future and the Past. Berkeley: University of California Press. ; Nippon Gakujutsu Shinkokai (1969). The Man'yōshū: The Nippon Gakujutsu Shinkokai Translation of One Thousand Poems. New York: Columbia University Press. Ponsonby-Fane, Richard. (1959). The Imperial House of Japan. Kyoto: Ponsonby Memorial Society. Titsingh, Isaac. (1834). Nihon Ōdai Ichiran; ou, Annales des empereurs du Japon. Paris: Royal Asiatic Society, Oriental Translation Fund of Great Britain and Ireland. Varley, H. Paul. (1980). Jinnō Shōtōki: A Chronicle of Gods and Sovereigns. New York: Columbia University Press. ; External links Asuka Historical National Government Park: image of Mausoleum Emperor Tenmu and Empress Jitō, exterior view Category:631 births Category:686 deaths Category:7th-century monarchs in Asia Category:7th-century Japanese monarchs Category:Japanese emperors Category:People of Asuka-period Japan Category:Buddhism in the Asuka period Category:Man'yō poets Category:Emperor Tenmu
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Bundesautobahn 999 was the designated number for the Mittlerer Ring in Munich, Germany, a projected second ring road around the inner city districts to accompany the Münchner Ring (Bundesautobahn 99). The project to construct the Mittlerer Ring as an autobahn was not carried out, instead the Bundesstraße 2R now covers most of the planned A 999 trajectory. History The urban development plan of 1963 provided for a ring road around Munich. This is referred to in the text only as outer trunk road ring, on the attached map, however, it is shown as a motorway ring. The planned motorway ring was planned much narrower than the later realized Federal Highway 99 and should run almost entirely on Munich city. Only in the south, the ring should be outside the city limits, the local course is therefore not specified in the urban development plan. According to the urban development plan, the ring on Munich's urban area should have the following course: from the BAB Salzburg (today A 8) around the planned relief settlement Perlach, along the peace promenade, west of Munich Riem and in the arc around Johanneskirchen over the Leintalerbrücke to the Nuremberg Motorway (today A 9), from there a bit north over Heidemannstraße and Rathenaustraße north on Fasaneriesee over to a part of a planned motorway ring in Allach and Langwied (Munich), already started before the Second World War, on the already existing Langwieder shamrock Crossing with the BAB Stuttgart (today A 8), continue to Aubing and the planned relief settlement Freiham along the city limits. Outside the city, the course in the arc near the city boundary should lead back to the BAB Salzburg (today A 8). Gallery External links 999 A999
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
2012 CONCACAF Men's Pre-Olympic Tournament qualification There will be a qualification held to determine the two qualifiers from the Caribbean and three from Central America who will join Canada, Mexico and the United States at the final tournament. Tiebreakers Per FIFA regulations the group tiebreakers for all qualifying tournaments will be: goal difference in all group matches greatest number of goals scored in all group matches If two or more teams are equal on the basis of the above criteria, their rankings shall be determined as follows: greater number of points obtained in all group matches between the teams concerned goal difference resulting from the group matches between the teams concerned greater number of goals scored in all group matches between the teams concerned play-off match on neutral ground (with extra time and penalty kicks, if necessary) Caribbean On 17 May 2011 CONCACAF announced the groups for the Caribbean qualifying. The group winners will advance to the second round, where two of them will qualify for the final qualifying tournament. First Round Group A Suriname was host for Group A. The Jamaica-Suriname match was abandoned in injury time with the score at 2–2 due to a pitch invasion and a brawl in which the Jamaican coach and players attacked officials and police; the result stands. Group B Dominica was host for Group B. Group C Cuba was host for Group C. Group D Saint Kitts and Nevis was host for Group D. Second Round St Kitts and Nevis was host for the Second Round. Central America On 3 May 2011, the UNCAF announced the two groups that will fight for three spots for the 2012 CONCACAF Men's Pre-Olympic Tournament. Group 1 Played from 13–17 September 2011 in El Salvador. Group 2 Played from 21–25 September 2011 in Honduras. Playoff Panama won 3–2 on aggregate. References External links 2012 CONCACAF Men's Olympic Qualification Regulations of the Olympic Football Tournaments London 2012 Qual
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Kosmos 686 Kosmos 686 ( meaning Cosmos 686), also known as DS-P1-Yu No.72, was a Soviet satellite which was launched in 1974 as part of the Dnepropetrovsk Sputnik programme. It was a spacecraft, which was built by the Yuzhnoye Design Bureau, and was used as a radar calibration target for anti-ballistic missile tests. A Kosmos-2I 63SM carrier rocket was used to launch Kosmos 686 from Site 133/1 of the Plesetsk Cosmodrome. The launch occurred at 16:34:56 UTC on 26 September 1974, and resulted in the satellite successfully reaching low Earth orbit. Upon reaching orbit, the satellite was assigned its Kosmos designation, and received the International Designator 1974-074A. The North American Aerospace Defense Command assigned it the catalogue number 07447. Kosmos 686 was the seventy-second of seventy nine DS-P1-Yu satellites to be launched, and the sixty-fifth of seventy two to successfully reach orbit. It was operated in an orbit with a perigee of , an apogee of , 70.9 degrees of inclination, and an orbital period of 91.8 minutes. It remained in orbit until it decayed and reentered the atmosphere on 1 May 1975. See also 1974 in spaceflight References Category:1974 in spaceflight Category:Kosmos satellites Category:Spacecraft launched in 1974 Category:Dnepropetrovsk Sputnik program
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Beaubassin East, New Brunswick Beaubassin East (2016 population: 6,376 ) is a Canadian Rural Community in Westmorland County, New Brunswick. The French form of the name is Beaubassin-Est. History Incorporation The RC was incorporated May 8, 1995 from the Local Service Districts of Boudreau West, Grand Barachois, Haute-Aboujagane, and Saint-André & LeBlanc Office, and portions of the parishes of Botsford, Sackville, and Shediac. It completely surrounds the Village of Cap-Pelé. Beaubassin East is divided into six wards, numbered counterclockwise from its eastern boundary. Ward 1 includes Bas-Cap-Pelé, Petit-Cap, Portage, and Shemogue Ward 2 includes Saint-André-LeBlanc Ward 3 includes Grand-Barachois Ward 4 includes Boudreau-Ouest Ward 5 includes Haute-Aboujagane Ward 6 includes Cormier-Village In 2006, Trois-Ruisseaux became part of Beaubassin East. Demographics Notable people See also List of rural communities in New Brunswick Greater Shediac References Category:Communities in Westmorland County, New Brunswick Category:Rural communities in New Brunswick
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
List of Huddersfield Town A.F.C. seasons This is a list of all the seasons played by Huddersfield Town Football Club in English football from 1908, when the club was formed. Seasons 32 seasons in Level 1 of English football league system 38 seasons in Level 2 of English football league system 22 seasons in Level 3 of English football league system 6 seasons in Level 4 of English football league system Key P = Played W = Games won D = Games drawn L = Games lost F = Goals for A = Goals against Pts = Points Pos = Final position NEL = North Eastern League ML = Midland Football League Div 1 = Football League First Division Div 2 = Football League Second Division Div 3 = Football League Third Division Div 4 = Football League Fourth Division Prem = Premier League Champ = EFL Championship Lge 1 = EFL League One 4QR = 4th Qualifying Round PR = Preliminary Round R1 = Round 1 R2 = Round 2 R3 = Round 3 R4 = Round 4 R5 = Round 5 QF = Quarter-finals NQF = Northern Quarter-finals SF = Semi-finals NSF = Northern Semi-finals NF = Northern Final R/U = Runners-up W = Winners Notes Seasons Huddersfield Town F.C.
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Falniowskia neglectissima Falniowskia neglectissima is a species of freshwater snail with an operculum, an aquatic gastropod mollusk in the family Hydrobiidae, the snouted freshwater snails. Distribution F. neglectissima occurs in Southern Poland and Ukraine. Habitat F. neglectissima inhabits freshwater springs. Conservation status F. neglectissima was originally assessed as Near Threatened (LR/nt) for the 1996 IUCN Red List, as it is restricted to springs in a small geographical area. However, individuals have not been found since 1988 and it is currently considered data deficient; it is listed as critically endangered in the Polish Red Data Book of Animals. References Category:Hydrobiidae Category:Gastropods described in 1989
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Raniera Te Aohou Ratana Raniera Te Aohou Rātana was the fifth president of the Rātana Established Church of New Zealand and the last surviving child of its founder Tahupōtiki Wiremu Rātana. He served the presidency from 1995 to 1998. He was succeeded by Harerangi Meihana (Harry Mason) in 1999. References Category:New Zealand religious leaders
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Dysmicohermes ingens Dysmicohermes ingens is a species of fishfly in the family Corydalidae. It is found in North America. References Further reading Category:Corydalidae Category:Articles created by Qbugbot Category:Insects described in 1954
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Rage of the Bloodbeast Rage of the Bloodbeast is Debauchery's second studio album. Track listing "Blood For The Bloodgod" "Chainsaw Masturbation" "Rage Of The Bloodbeast" "Devour Of Worlds" "Killing In The Warzone" "I Will Rape And Murder" "Armies Of Immortals" "The Hellspawn" "Ritual Killing" "Wargrinder" "Take My Pain" Personnel Thomas - vocals, guitars, producer Dani - drums, co-producer Guest musicians Jens Müller - piano Sandra Hoffmann - female victim on "Chainsaw Masturbation" Marie and Hanna Fleig, Bettina Kruck, Carmen Farian - string quartet External links Debauchery's website Category:2004 albums Category:Debauchery (band) albums
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Birte Høeg Brask Birte Høeg Brask nicknamed Trille (1918–1997) was a Danish resistance fighter and physician. During the German occupation of Denmark in World War II, she became a member of the Danish resistance. As a communist, after Germany attacked the Soviet Union in 1941, together with her husband Kjartan Munck, she contributed to the first clandestine publications in Denmark. She went on to collaborate with the writer and resistance fighter Børge Houmann, working as a courier, organizing illegal meetings and contributing to the resistance papers Land og Folk and Ugens Nyt. Continuing her studies after the war, she became a specialist at the children's psychiatric hospital in Aarhus. From 1970 until her retirement in 1988, she held the position of chief consultant. Høeg Brask is remembered in particular for her ground-breaking contributions to autism. Early life and family Born in the Copenhagen district of Frederiksberg on 23 November 1918, Birte Høeg Brask was the daughter of the laboratory director Axel Høeg Brask (1875–1942) and Emma Brincker (1879-1968). The youngest of three sisters, she was brought up in the Vesterbro district. While a student at Falkonergårdens Gymnasium, she became active in the left-wing Dansk Gymnasiastforbund (Danish High School Association). After matriculating in 1936, she studied medicine at the University of Copenhagen where she met Kjartan Munck, a communist. They married in 1937 and had a son, Ole, in 1939. In 1946, her marriage with Munck was dissolved. She became the life-long partner of Børge Kruuse Houmann (1902–1994) whom she married in 1962. Resistance work Nicknamed Trille, Høeg Brask showed interest in social and cultural radicalism from an early age. As a result, she was attracted to participating in illegal activities during the German occupation. After the German invasion of the Soviet Union, she became firmly committed to the resistance cause in the autumn of 1941. Together with her husband, she became involved in the distribution of Danske Toner, the first illegal paper in Denmark which was published by Børge Houmann, one of the early leaders and later head of the communist resistance movement. Using the cover name Mads, she became one of Houmann's closest associates for the remainder of the German occupation, organizing hiding and meeting places, clandestine printing operations and performing extensive courier work. She became a major contributor to the illegal publications Land og Folk and Ugens Nyt, based on information received from Kremlin radio broadcasts. During the war, Høeg Brask put her son Ole into care in order to devote all her efforts to the resistance. She later regretted this decision. Post-war career After the war, Høeg Brask completed her medical studies, graduating in 1946. She became increasingly interested in pediatric psychiatry, becoming a specialist in 1957. From 1958, she headed the psychotic children's department in the newly established child psychiatric hospital in Aarhus. From 1970, she served as the hospital's chief consultant until her retirement in 1970. One of her colleagues, Vandborg Sørensen, commented: "As a human being, she was a fascinating person... Professionally, she was inspirational, because her insight into children with autism was so impressive." In 1988, an account of her work with psychotic children was published as Tilegnet Birte Høeg Brask: et skrift om psykotiske børn. Birte Høeg Brask died in Aarhus on 1 January 1997. References Category:1918 births Category:1997 deaths Category:People from Frederiksberg Category:Danish resistance members Category:Danish female resistance members Category:Danish psychiatrists Category:Danish women physicians Category:20th-century Danish writers
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Jeholochelys Jeholochelys is an extinct genus of sinemydid turtle that lived during the Early Cretaceous of what is now China. The holotype specimen was discovered in the Jiufotang Formation of Sihedang in Lingyuan, western Liaoning. In 2018, the Chinese palaeontologist Shuai Shao and colleagues named the new genus and species Jeholochelys lingyuanensis based on the specimen. The generic name consists of "Jehol", which refers to the Jehol biota, and "chelys", which is Greek for turtle. The specific name refers to the type locality. Seven skeletons were described in the study, five nearly complete, and two consisting of shells (four appear to be juveniles), and hundreds of turtle fossils have been found in the area. The described specimens are kept in the Paleontological Museum of Liaoning. Jeholochelys lived in freshwater, and was characterised by hyperphalangy, the increase in the number of phalanx bones in the digits (it had one additional phalanx bone in the fifth toe compared to what is common among living turtles). This condition is often linked to tetrapod animals with an aquatic lifestyle, where it contributes in forming long flippers in for example modern whales, as well as in the extinct ichthyosaurs and plesiosaurs. Modern soft-shelled turtles also exhibit hyperphalangy, but though this does not result in long flippers, it may help enlarge the paddle surface and aid aquatic movement. Other aquatic and marine turtles instead have elongated limbs and phalanges, and land turtles have short limbs and feet. Jeholochelys was found to be related to coexisting sinemydid turtles, most closely to Xiaochelys and Changmachelys, and like other Cretaceous relatives, it had a low-domed shell. Jeholochelys fossils are considered evidence that hyperphalangy evolved multiple times among turtles; other coexisting sinemydids did not have an additional phalanx bone. The proportions of the forelimb in Jeholochelys was similar to that in soft-shelled turtles, and it may have been adapted for aquatic habits to a similar degree. References Category:Cryptodira Category:Early Cretaceous turtles Category:Early Cretaceous reptiles of Asia Category:Fossil taxa described in 2018 Category:Prehistoric turtle genera Category:Extinct turtles
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
All Saints Church, Fleet All Saints' Church is the Anglican parish church of the town of Fleet in the county of Hampshire, England. It is notable for its architect, William Burges and was constructed 1861-2. History and description The church was designed by Burges and was built between 1861 and 1862. It was extended to the west in 1934 by A. J. Steadman and a Lady Chapel was added in 1958 by John Purser. The church was commissioned by the local squire, Charles Edward Lefroy, secretary to the Speaker of the House of Commons as a memorial to his wife, Janet Lefroy. It cost £3,323. It is a Grade II* Listed Building Pevsner says of Fleet that "it has no shape, nor character nor notable buildings, except one," that one being All Saints. The church is of red brick and its plan is "simple in the extreme." The interior too is simply decorated but the massive sculpture, particularly of the Lefroys' tomb and of the gabled arch below which it originally sat is quintessentially Burges, "not so much muscular (gothic) as muscle-bound." Janet Lefroy was originally a Walker, daughter of James Walker, who established the great marine engineering company of Walker and Burges with Alfred Burges, father of William. This family connection led to Burges obtaining the commission. The interior was significantly altered in the later twentieth century by the removal of the Lefroy tomb from its original position and by the whitewashing of the apse. Of these alterations, Pevsner comments; "And so a work by a none too prolific genius is irreparably spoiled." Fire Far greater damage was caused in 2015 by an arson attack. On 22 June the roof and much of the interior was destroyed by a major fire. Daniel Finnerty, a seventeen-year-old local, was subsequently arrested for suspected arson. In November 2015, at Winchester Crown Court, Finnerty pleaded guilty to two charges of arson and was sentenced to four years in a Youth Offender Institution (reduced to years on appeal) and to a three-year extended licence. Notes References Pevsner, Nicholas and Lloyd, David, Hampshire and the Isle of Wight, The Buildings of England, Penguin (1967) Mordaunt Crook, J, William Burges and the High Victorian Dream, James Murray (1981) External links Category:Churches completed in 1862 Category:William Burges church buildings Fleet, All Saints Category:Gothic Revival church buildings in England Category:Gothic Revival architecture in Hampshire Fleet Category:1862 establishments in England Category:Fleet, Hart
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Svetlana Kapanina Svetlana Vladimirovna Kapanina () is a Russian aerobatic pilot. Biography Kapanina was born on 28 December 1968 in Shchuchinsk, Kazakh SSR, Soviet Union (now Kazakhstan). She dedicated herself to several sports modalities at school and always liked motorcycles and other motor vehicles. She enrolled at medical school in Tselinograd (now Nur-Sultan), where she graduated in pharmaceutical sciences. She started flying at age 19, in 1988, on a Sukhoi Su-26M3, while working as a technician at the Kurgan sports aviation club of DOSAAF. By 1991 she was already an instructor pilot at DOSAAF's Irkutsk club, and then back at Kurgan. Also in 1991, she became a member of the Russian national aerobatic team. In 1995 she graduated from Kaluga aeronautical technical school. She lives in Moscow with her husband and two children. Achievements in aerobatics Kapanina was World Aerobatic Champion in the women's category in 1996, 1998, 2001, 2003, 2005, 2007 and 2011 and has won the title more times than any other pilot in the category. Additionally, she was overall World Air Games Champion in 1997 and 2001. Together with Mikhail Mamistov and Oleg Spolyansky, she won the team gold medal in the 16th FAI European Aerobatic Championships 2008 in Hradec Králové (Czech Republic). She placed fourth overall and was best female participant. Awards In 1997, she received the Paul Tissandier Diploma by the Fédération Aéronautique Internationale (FAI). In 2005 she was awarded the Sabiha Gökçen Medal and the Centenary Medal by the FAI. She was awarded Russia's Order of Courage by President Vladimir Putin on 22 December 2014. See also Competition aerobatics FAI World Aerobatic Championships FAI European Aerobatic Championships References External links Official website Category:Aerobatic pilots Category:Russian aviators Category:People from Akmola Region Category:Sportspeople from Moscow Category:1968 births Category:Living people Category:Russian female aviators
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
ES Souk Ahras Entente sportive of Souk Ahras (), known as ES Souk Ahras or simply ESSA for short, is a football club based in the city of Souk Ahras, founded on 2,November 1957. . History History of the club Entente Sportive of Souk-Ahras: The word agreement used during the French colonial in order to make hear between French and Arabs. The first official match against the team of Oum El Bouaghi whose agreement to win the match (2-0), was the beast of the club of the East of Algeria and is also called to this time the (Red Devil). At independence takes the name (Fanal Red). Created on 2 November 1957, he wears the flag of the city in several sports, including football, where he participates in the regional league of Annaba, and the football team of the 1970s when the (ESSA) evolved in 2nd division and gained Several game against big national clubs. This club is considered among the oldest club of Algeria and that spawned great players that participated in the national team. Example: Yacine SID. The different names of the club 1980 Rapid Town of Souk-Ahras (RCSA) Colours and badge For the red color, it is a flame that is the heart of the players and the blood of the martyr and the eternal love of the club. The black is the color symbolizes power. Lion of Barbary symbol of the city. Stadium Badji Mokhtar of Souk Ahras, football stadium in the town of Souk Ahras, the stadium can accommodate 15,000 spectators with natural grass lawn, it has an athletics tartan track Annexes likely to host high level competitions. Supporters and rivalries The club ESSA is the most popular club in Souk Ahras, its supporters are called lions, literally in Arabic Algeria (السبوعة), in reference to their number. Among the fan groups of the club, the Ultras (Red Lion), he is the first group of Ultras in Souk Ahras. References External links website of fans A newspaper article about the club A newspaper article about the club A newspaper article about the club Interview of Chouchane Salah ES SOUK AHRAS A newspaper article about the club Category:Football clubs in Algeria Category:Association football clubs established in 1957
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Chadian intervention in northern Mali On 18 January 2013, the Republic of Chad announced its intent to deploy 2,000 troops: one infantry regiment with 1,200 soldiers and two support battalions with 800 soldiers, into Mali as part of the international campaign against Islamist insurgents. The Chadian forces were not part of the African-led International Support Mission to Mali but integrated into an existing French command structure. The same day the arrival of the Chadian army in Niamey was confirmed. Backed by technicals and Eland-90 armoured cars, Chadian forces entered Mali via Niger, securing the former Islamist town of Ménaka on 28 January. According to a statement by the French Ministry of Defense, Chadian forces left their base in Ménaka and headed towards the north of the country in support of Malian armed forces in the area. On 31 January, some 1,800 Chadian soldiers entered the city of Kidal without resistance, as French forces captured the outskirts of the city a few days prior to the formal capture. The Chadian army contributed to the security of the city, currently held by the MNLA. On 7 February, in the evening Chadian and French forces entered the town of Aguelhok. On 8 February, French and Chadian forces announced they captured the town of Tessalit near the Algerian border, the seat of one of the last airports still held by insurgents. On 12 February, according to numerous sources, a Chadian soldier succumbed to an illness, marking the first death since the Chadian intervention began in late January. On 22 February, supported by French fighters, the Chadian army launched a joint military operation with the support of French war jets on an Islamists base said to be of "significant importance", where 8 French hostages were believed to be held along with caches of heavy weaponry, artillery shells, anti-tank weapons, and mines were said to be hidden in the mountains of the Adrar des Ifoghas. By the end of the day, the mission was considered to be a success with 93 Islamists killed or captured, but leaving 26 Chadian soldiers dead and 52 others severely wounded, the dead included several high-ranking Chadian commanders in Mali, including special forces commander Abdel Aziz Hassane Adam. On 25 February, during a joint military operation in the Adrar des Ifoghas mountains, Chadian forces reported the killing of al-Qaeda right-hand leader of AQIM, Abou Zeïd, along with 40 of his followers. It's unknown whether he was killed by French airstrikes or by Chadian ground forces, as the information was neither confirmed nor denied by the French Government. On 1 March, the Chadian army claimed to have killed Mokhtar Belmokhtar. According to a government statement, Mokhtar along with several other extremists were killed, and weapons, equipment, and 40 vehicles were seized. In April, the Chadian government dismissed their previous claim of killing Mokhtar, saying he blew himself up, along with several other militants, in despair after learning about the death of Abou Zeïd. On 12 March, a Chadian soldier was killed in a skirmish along with six Islamists, according to military officials. On 12 April, a suicide bomber detonated his explosive belts while a group of Chadian soldiers were passing by in a busy market in Kidal, killing four soldiers and the bomber in the blast, which also injured five civilians. On 14 April, Chadian president Idriss Déby Itno announced the full withdrawal of Chadian soldiers from Mali, citing face to face with the Islamists is over and the Chadian army doesn't have the skills to fight a guerilla-style war, referring to the previous attack in Kidal that killed four of its
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
soldiers. According to local sources the army had already withdrawn a mechanized battalion a few days prior to the formal announcement and pulled out its troops from Aguelhok and Tessalit. On 15 April, the Chadian parliament voted overwhelmingly for a resolution for the withdrawal of Chadian forces in Mali, "within a reasonable timeframe". The same day the Chadian army announced that it lost 36 soldiers during the three-month intervention in northern Mali and suffered another 74 wounded. In total the presence of Chadian soldiers in Mali costs the country 57 billion CFA francs (87 million). During the intervention, at least 2,250 soldiers were deployed with 250 vehicles and a heavy amount of supplies. On 17 April, tension grew between the separatists Tuareg group MNLA and the Chadian army stationed in Kidal. The MNLA accused the Chadian army of planning to allow Malian forces to be deployed to the city and firing on unarmed civilians during demonstrations against Chadian military rule. References Category:2013 in Mali Category:Wars involving Chad Category:Mali War
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Khajurgachhi Khajurgachhi is a village development committee in Jhapa District in the Mechi Zone of south-eastern Nepal. At the time of the 2001 Nepal census it had a population of 7471 people living in 1544 individual households. References Category:Populated places in Jhapa District
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Brothers (Ugly Betty) "Brothers" is the 15th episode from the dramedy series Ugly Betty, which aired on February 8, 2007. Plot In the three days since the bombshell at Fashion Week, Alexis Meade's shocking announcement has left MODE in chaos and the dysfunctional Meade family at each other’s throats. As the day begins, Betty has to explain to a curious Ignacio about the difference between a Transvestite and Transsexual after he comments on how small Alexis' hands are. This is followed by Betty finally meeting Alexis in the elevator at work, where after Betty gives her a full 'look over' Alexis tells her that she had everything done...and that despite a rumor mentioned on Fashion TV, no she did not keep 'it' in a jar. As Betty tries to talk to Daniel about Alexis, she gets no comment from him; all he wants to do is focus on MODE's upcoming swimwear issue. But as he begins the staff-only meeting Alexis shows up to make massive changes, by firing Daniel and placing Wilhelmina in charge of MODE as the new editor-in-chief. Daniel tells Betty that they aren't going anywhere; he has an injunction because Alexis had a sex-change, and therefore cannot take over Meade Publications since Alexander Spencer Meade is legally dead. And to make matters worse, Claire is excited to see her son-turned-daughter back, which is not sitting well with Daniel, thus prompting Betty to step in by taking Daniel's mother to a spa to calm things down and try to find out about how to squash the Meade squabble, and keep an eye on Claire in case she has a nervous breakdown. Betty gets nervous for her first time at the spa. She starts with a mud bath and then she and Claire head over to the massage table. On the massage tables Claire is having a relaxing facial by a dark hair haired female beautician and Betty is having a painful massage by a large masseuse who is repeatedly hitting her with tapotement bamboo sticks and bending her legs and pulling her arms back. Betty is screaming "OW!" and finding it too difficult to relax. The sibling rivalry between the Meades heat up a lot further when Alexis tries to hold a press conference, and Daniel interrupts it by taunting Alexis and telling the crowd that Alex wet the bed until he was 15 (which was his response after Alexis brought up his brother's 'romance' with Sofia Reyes). This leads to a fight between the two in front of the camera, which is put to a halt by Betty. She then talks the two into going to their childhood getaway, Coney Island, after realizing that Claire is there. When the two see their mother in a bumper car, they end up getting into separate bumper cars, and Betty makes sure that all three don't leave by getting into a bumper car herself until they bury the hatchet. Daniel expresses his deep anger at Alexis for letting the family think (s)he was dead, and also that she didn't tell Daniel she was going to have a sex-change. She gets very emotional and explains that she had confided in Bradford about it (which further upsets Daniel) and he threatened to disown Alex if it happened. Alexis then states that she couldn't stand being rejected like that twice, so that's why she never told Daniel. Claire remarks to Betty that maybe it was not a good idea to trick the siblings into talking. Reality sets in the following day for Alexis and Daniel, who both agree that they should work out
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
their differences by working together. Later that same day, Wilhelmina, upset that her new role as EIC is now short-lived because of the injunction, plots her next move by telling Alexis that she won't set foot in MODE until the injunction is resolved. She also 'meets' her housekeeper, Evelyn, who has actually been working for Wilhelmina for thirteen years, but whom Wilhelmina has never seen as the housekeeper works during the day. We also learn that Wilhelmina's walkout is actually a front; at home she calls Marc to tell him to check on his flowers and tells him to rotate them in front of him. As he does, she tells Marc to check his nose hairs; she can see everything from the flowers, which is actually a hidden camera that can be seen via Wilhelmina's laptop computer. She also gets a glimpse of people celebrating Wilhelmina's departure (but Marc lies, telling her it's a birthday party). The camera comes in handy as she watches Daniel and Alexis shake hands and be civil, leading Wilhelmina to suspect that she has once again been screwed over, this time by both Meades. Meanwhile, Betty is in for even more turmoil in her love life, as Henry asks her out while Walter tells her he’s got a big surprise in store. At work Henry asks Betty if she wants to go to see Wicked after learning that it is her favorite musical, but Betty says that they'll go as friends because of her relationship with Walter, which disappoints Henry. Speaking of Walter, it turns out that he has accepted a job offer as the assistant manager at a new Pro-Buy in Maryland. Knowing that he and Betty are drifting apart and wasn't meant to be a couple any longer, Walter tells her that he is taking the job, and they have their final kiss. Betty sincerely tells Walter she'll miss him. The next day the now-single Betty gets a shock from Henry when he tells Betty that he went back to Arizona over the holidays, and is going to try to work things out with his ex-girlfriend, who just showed up in New York by surprise. He also gives Betty the tickets, thinking that she is still seeing Walter. Back at home, Hilda is miffed when Santos gets involved after Justin is sent home from school for fighting. Hilda wonders why Justin would call Santos instead of her, but Santos says Justin has every right to call his dad. Hilda learns from a visiting mother that Justin hit her son during flag practice because the boy made a comment about Hilda being a "slut mom" (presumably in reference to her mode of dress). Both Hilda and Santos defend Justin's actions. Justin asks Hilda if she's mad at him, and she tells him she's not mad at him at all. Santos praises Justin, saying he stood up for his mom and he's proud of Justin. The following day Justin sees Hilda trying to be dress conservatively, but Justin says he accepts Hilda for who she is and how she dresses, even if she does overaccessorize. She's always telling him to be proud of who he is and not let anyone make him feel bad about it, so shouldn't she do the same thing? Claire appears to have booked herself into a hotel, and run up so many charges that the magnetic strip has worn out. Daniel gives Betty a credit card to take to Claire. When Betty gets to Claire's room, she sees Claire is drinking again and then is offered a drink (at 10am).
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Betty assures her that everything will be all right and that Bradford is not a killer. Claire replies to Betty "I know who killed Fey Sommers and it wasn't Bradford. It was me." After hearing that confession, Betty swallows her martini in shock. Production The episode's original working title was "Brothers and Friends", which was the episode's title that aired in Hong Kong. On iTunes, that is also the episodes listed name. This would also be Kevin Sussman's last appearance in this episode as Walter, even though Sussman's name would continue to be listed in the series as a regular throughout the remaining seven episodes. Also, when Amanda was caught faxing her resume by Wilhelmina, she tears a piece of her resume and puts it in her mouth. As she was chewing on it her resume that she was holding in her hand was not torn. But when she said "amen" her resume was torn again. There are two deleted scenes that were cut from the episode, an extended story on Alexis' return and Wilhelmina's closetful of dresses at her apartment. The scenes are featured in the extras CD on the season one DVD. In one scene Marc tells Wilhelmina that Anna Wintour called, but Wilhelmina says that she should've returned her call back in 1998 instead of 2007. Wintour was of course the inspiration for both series characters Wilhelmina Slater and Fey Sommers. Also, Wilhelmina's maid, Evelyn, watches the same telenovela that the Suarezes watch, "Muchas Muchachas," whose segment was featured in the episode. | image = | caption = Betty receives a painful massage by a large masseuse. Casting Ashley Jensen, although credited, was absent from this episode. Other notes In this episode we discover Meade Publications actual address when Wilhelmina picks up one of her business cards: 100 MEADE PLAZA SOUTH, NEW YORK, NY 10016'''. Accounting is on floor 3, Mode on floor 28, MYW on floor 27, Psychology Now (Mentioned in "Secretaries' Day") on floor 14, and owner/CEO Bradford Meade's office on the Penthouse, which is the 30th floor. Another piece of Meade family history was revealed by Claire when she mentions that Daniel lost one of his pinky toes after Alex pushed him into an icy lake. According to Hilda, Justin was conceived in the back seat of Santos's rented Camaro. Santos later hints that this occurred on Prom Night. Hilda confirms this on episode 18 during dinner with Mrs. Weiner and Mark when she says that the condom broke. This was later confirmed in episode 20 when Hilda tells Justin how she got a "precious reminder" of her prom night 9 months after it happened. The second episode which Betty has a massage. The first being After Hours'' which she has a relaxing one by a guy. And a painful one in this episode by a large woman. Betty still wears her glasses when she's having a mud bath and having a massage. Reception The episode received mixed reviews because of its drama-heavy theme throughout the show. Notes Entertainment Weekly's Tanner Stransky, "I suppose that by definition a dramedy will be a mix of funny and heavy moments. After the past couple of episodes of Ugly Betty — which focused on lighthearted topics like fashion shows and swag — this one hit me a little hard, falling much more in the drama column. Bradford is incarcerated, Daniel and Alexis duked it out, Betty avoided Walter, Justin got in a fight at school, Claire admitted to killing Fey Sommers. Ughhh, why don't you just stab me with a Christian Louboutin spike, UB! Or suffocate me
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
with a Kate Spade. The laughs are just so good it sometimes makes the drama a harder pill to swallow." Awards "Brothers" was submitted for consideration in the category of "Outstanding Writing in a Comedy Series" for the 2007 Emmy Awards. In addition, Judith Light submitted this episode for consideration of her work in the category of "Outstanding Guest Actress in a Comedy Series". References Also starring Judith Light (Claire Meade) Kevin Sussman (Walter) Christopher Gorham (Henry) Kevin Alejandro (Santos) unknown actress as Female Beautician unknown actress as Large Masseuse Guest stars Tim Gunn (himself, as "Fashion TV" reporter) Liz Torres (Evelyn) Funmi Desalu (Assistant #2) Ogy Durham (Reporter #3) Lauren Glenn (Assistant #3) Jeffrey Scott Kelly (Guy) Jill Latiano (Assistant #1) Abigail Marlowe (Reporter #1) Amy Lyndon (Shelly) Jack Maxwell (Reporter #2) External links Category:Ugly Betty (season 1) episodes Category:2007 American television episodes
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Eric Smith (British politician) Eric Martin Smith (28 December 1908 – 13 August 1951) was a British Conservative Party politician. He was a noted amateur golfer, winning the Amateur Championship in 1931 and representing England against Scotland and Ireland the same year. Smith was the son of Everard Martin Smith and Violet, the daughter of Everard Hambro. He was educated at Eton College and Cambridge University. At the 1950 general election, he was returned as the Member of Parliament (MP) for the Grantham constituency in Lincolnshire. He replaced the independent MP, Denis Kendall, who had been first elected in a war-time by-election, but came third in 1950. Smith died in August 1951, aged 42, of a heart attack, at his home in Hitchin, Hertfordshire. His death did not trigger a by-election. The seat remained vacant until the 1951 general election, on 25 October. References External links Category:1908 births Category:1951 deaths Category:People educated at Eton College Category:Alumni of the University of Cambridge Category:Conservative Party (UK) MPs for English constituencies Category:UK MPs 1950–1951 Category:Politics of Grantham Category:English male golfers Category:Amateur golfers
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Omobono Stradivari Omobono Stradivari (November 14, 1679June 8, 1742) was a violin maker and the sixth son of the famous Antonio Stradivari. Biography Omobono Stradivari was born on 14 November 1679 to Antonio Stradivari and his first wife Francesca Ferraboschi in Cremona, Italy. Little is known about his early life except that his family was living at a house known as Casa del Pescatore in the Parish of Saint Agata. Just one year after Omobono was born, the family sold the Casa del Pescatore for the respectable sum of 7,000 Imperial Lire, and relocated to the famous home of Antonio Stradivari, Casa Nuziale, which is now No. 1 Piazza Roma, Cremona. When he was 19, his mother died and his father Antonio organized an elaborate funeral, showing that the Stradivaris were already quite prosperous. Antonio quickly remarried, this time to Antonia Zambelli, and had five more children. While it is probable that Omobono started working in his father's workshop from a young age, as was the practice of the day, there is very little work that can be definitively traced to him. Antonio Stradivari is well known not only for his high productivity, but also for his use of templates and a highly finished product, eliminating all tool marks that could be used to deduct the identity of the workman. Omobono is supposed to have been chiefly occupied with repairing. On works that are attributable to him, he shows clumsy workmanship, except on the scroll, suggesting that if his father employed a division of labour in his workshop, perhaps Omobono was assigned the scrolls. Besides the decay in workmanship, Omobono's instruments show little other departure from the design of his father except that he tended to point the mitres of his purfling outward, away from the Cs. After his father's death in 1737, Omobono, who was evidently enriched by his inheritance, stopped his production of instruments, and receded into obscurity. References Category:1679 births Category:1742 deaths Category:People from Cremona Category:Bowed string instrument makers Category:Italian musical instrument makers
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Marco Taradash Marco Taradash (born 19 May 1950, in Livorno) is an Italian politician and journalist. Biography Marco Taradash was born in Livorno on 19 May 1950 from a Tuscan mother and a New Yorker father, who came to Italy with the allied army. The surname has a Ukrainian Jewish origin, but while the grandmother, Zweig, kept the Jewish faith, the paternal grandfather, who fled as a child with the family to the United States, later converted to Catholicism, and the father to Protestantism. In 1968 he enrolled in the youth organization of the Italian Liberal Party, which he left in the mid-seventies, to join the Radical Party. In the 70s he also began his journalistic career and thanks to "Stampa e Regime", the press review of Radio Radicale, he also obtained one of the most famous Italian journalistic prizes, the Premiolino. He was elected to the European Parliament in 1989 for the Antiprohibitionists on Drugs list, while in 1994 and in 1996 he was elected to the Chamber of Deputies for Forza Italia. He also served as Chairman of the RAI Supervision Commission from 1994 to 1996. In the 2000s he conducts some television programs of political depth. In 2005 he founded, along with Peppino Calderisi and Benedetto Della Vedova, the Liberal Reformers movement. In 2009 he was a candidate for mayor of Livorno with the support of the centre-right coalition but he was defeated by the outgoing mayor Alessandro Cosimi, candidate for the Democratic Party. In 2010 he was elected to the Regional Council of Tuscany for The People of Freedom, but in 2013 he left it to join the New Centre-Right, a party led by Angelino Alfano. In the European elections of 2019 he is a candidate for the European Parliament with More Europe, in the central Italian constituency. References Category:1955 births Category:Living people Category:People from Livorno Category:Italian journalists Category:20th-century Italian politicians Category:21st-century Italian politicians Category:Radical Party (Italy) politicians Category:Forza Italia politicians Category:The People of Freedom politicians Category:New Centre-Right politicians
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Wāḳwāḳ Wāḳwāḳ (Wāḳ Wāḳ, Wāḳ al-Wāḳ, al-Wāḳwāḳ; Arabic: ) is the name of an island, or possibly more than one island, in medieval Arabic geographical and imaginative literature. And in the Arab versions, in the sea of China there is also the famous island of Waq-Waq as reported by most medieval Arab geographers. This island is ruled by a queen and the population is only female: it is usually illustrated in al-Qazvini manuscripts of the Wonders of Creation showing the queen surrounded by her female attendants. Ibn Khordadbeh mentions Waqwaq twice: "East of China are the lands of Waqwaq, which are so rich in gold that the inhabitants make the chains for their dogs and the collars for their monkeys of this metal. They manufacture tunics woven with gold. Excellent ebony wood is found there. And again: Gold and ebony are exported from Waqwaq." Michael Jan de Goeje offered an etymology that interpreted it as a rendering of a Cantonese name for Japan. Gabriel Ferrand identified it with Madagascar, Sumatra or Indonesia. Wāḳwāḳ is referred to in a number of sources; it is generally an island far away. The waqwaq tree Tongdian by Du Huan mentions an Arab account of a tree growing little children. In the Kitab al-Bulhan, the painting titled the ‘Tree of Waq Waq’ is rather extraordinary because it illustrates the way in which the all-female population reproduces and self-perpetuates. Female figures grow from the tree as if they mature like fruit until they are ‘ready’ and they drop to the ground emitting a cry that sounds like ‘Waq Waq!’ An Andalusi versions mentions beautiful women as the fruit of the tree. See also Jinmenju, a strange Japanese tree in Toriyama Sekien's Konjaku Hyakki Shūi. Nariphon, a tree in Buddhist mythology which bears fruit in the shape of young female creatures. Zaqqum, a tree in Jahannam, the fruits of which are shaped like the heads of devils. References Category:Historical regions Category:Islands of the Arabian Sea Category:Arabian mythology Category:Trees in mythology Category:Mythological islands
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Witkówko Witkówko () is a village in the administrative district of Gmina Witkowo, within Gniezno County, Greater Poland Voivodeship, in west-central Poland. It lies approximately north-west of Witkowo, south-east of Gniezno, and east of the regional capital Poznań. The village has a population of 60. References Category:Villages in Gniezno County
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Yeh Ishq Haaye Yeh Ishq Haaye is an Indian television series that aired on STAR One. The show premiered on 1 November 2010 and ran till 6 August 2011. Plot Yeh Ishq Haaye is set in small town Agra and traces the lives of childhood friends Manjari and Akshay. Manjari Chaturvedi is a carefree teenager who likes to hang out with her friends. Akshay runs a cafe and has secretly been in love with Majari for a long time. Pooja, Manjari's elder sister, falls for Akshay after a misunderstanding leads her to believe that he returns her feelings. Manjari instead falls in love with Ranbir. Akshay, on Majari's insistence gets engaged to Pooja. Tired of her carefree attitude, Manjari's parents want to get her married so she settles down but Manjari decides to run away to Mumbai on Ranbir's advice. Akshay insists on accompanying her and on reaching Mumbai, they realise that Ranbir is in fact Sanjay and he cheats girls using a fake identity. Akshay and Manjari later decide to stay in Mumbai and make a name for themselves. Manjari starts to fall in love with Akshay but stays away from him as he is Pooja's fiancee. Back in Agra, Pooja comes to know of Akshay's feelings for Manjari. She comes to Mumbai and plans to separate them but later realises their love and leaves. Fighting against all odds, Manjari and Akshay unite at last. Music Title theme song was sung by Alisha Chinai & Mohit Pathak and penned down by Amitabh Bhattacharya. Cast Srishty Rode as Manjari Neha Desai as Pooja, Manjari's elder sister Raunaq Ahuja as Akshay Shukla, Manjari's childhood friend who has been in love with her Jaya Ojha as Manjari's mother Pankaj Berry as Manjari's father Rahul Arora as Sanjay Sharma Anand Suryavanshi as Ranbir Malhotra Yashdeep Nain as Ajay Shukla, Akshay's brother References External links http://www.starone.in/show.aspx?sid=75 http://uk.startv.com/?q=synopsis/yeh-ishq-haaye http://www.starone.in/news/840/latest-offerings-from-star-one.aspx Category:STAR One television series Category:2010 Indian television series debuts Category:2011 Indian television series endings Category:2010s Indian television series Category:Indian drama television series
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Jan Lee Jan L. Lee (born May 30, 1943) is an American politician who served in the Oregon House of Representatives from 2001 until 2003. Biography Lee was born in Santa Barbara, California in 1943. She graduated from Linfield College in 1982, and received a Master of Public Administration from Lewis & Clark College in 1989. Lee served on the board of the Clackamas Soil and Water Conservation District from 1999 until 2000. She was elected to the Oregon House in 2000 as a Republican, but switched to become an Independent in June 2001. Regarding her switch, she stated, "There is still strong partisanship that stands in the way of bringing people together in the middle to form workable solutions. You cannot force someone to follow hard-line conservative beliefs that do not serve their community and do not reflect their values." She later switched to the Democratic party, and ran as a Democrat for the 51st district in the 2002 elections, where she was defeated by Linda Flores. Personal life Lee is divorced, and has 2 children. She resides in Oregon City. See also List of American politicians who switched parties in office References Category:1943 births Category:Living people Category:Lewis & Clark College alumni Category:Linfield College alumni Category:Members of the Oregon House of Representatives Category:Oregon Democrats Category:Oregon Independents Category:Oregon Republicans Category:Politicians from Oregon City, Oregon Category:People from Santa Barbara, California Category:Women state legislators in Oregon Category:American Lutherans Category:21st-century American politicians Category:21st-century American women politicians
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Jacob Lyda Jacob Lyda (born May 18, 1975) is an American country music singer. His first four albums were produced by Mike Headrick for BGR Records/Country Discovery Records 1995-2003. He is now signed to Davis Music Group and is being promoted by Nine North Records. In early 2011, Lyda released his debut single "I'm Doin' Alright", which has charted on Hot Country Songs. Lyda co-wrote the song with Paul Overstreet. The song was to have been included on his album Another Song I Just Had to Write. Karlie Justus of Engine 145 gave the song a "thumbs up", saying that the song was "something greater than the sum of its parts". She praised its sound and the "charisma" in Lyda's voice. Discography Albums 1995 - My Dreams Just Came True - BGR 9502 1997 - Here We Go Again - BGR 9509 1998 - Three Times the Charm - BGR/Country Discovery CD1103 2003 - Back in the Swing - Country Discovery CD1052 Singles Music videos References External links Category:American country singer-songwriters Category:People from Stevenson, Alabama Category:American male singer-songwriters Category:Living people Category:1975 births Category:Singers from Alabama Category:Songwriters from Alabama Category:21st-century American singers Category:Country musicians from Alabama Category:21st-century male singers
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Judgement of Martin Bucer Concerning Divorce Judgement of Martin Bucer by John Milton was published on 15 July 1644. The work consists mostly of Milton's translations of pro-divorce arguments from Martin Bucer's De Regno Christi. By finding support for his views among orthodox writers, Milton hoped to sway the members of Parliament Protestant ministers who had condemned him. Background Milton married in Spring 1642, and shortly after, his wife Marie Powell, left him and returned to live with her mother. The legal statutes of England did not allow for Milton to apply for a divorce and he resorted to promoting the lawfulness of divorce. Although the laws did not change, he wrote four tracts on the topic of divorce, with Judgement of Martin Bucer Concerning Divorce as his second tract. The hostile response by clergymen to the first tract, The Doctrine and Discipline of Divorce, prompted Milton to defend himself by translating Martin Bucer's De Regno Christi and his arguments concerning the legitimacy of divorce. Bucer was a Protestant Reformer and close to the Protestant movement in England, and Milton felt that he would serve as a means to convince Parliament to change their views on divorce. The work was published on 13 August 1644, a week before Milton was attacked in a sermon preached before Parliament by Herbert Palmer. Tract The work begins with a preface titled "To the Parlament", and the preface connects the history of Bucer and his reformist ideas with the history of Milton's previous tract on divorce: For against these my adversaries, who before the examining of a propound truth in a fit time of reformation, have had the conscience to oppose naught els but their blind reproaches and surmises, that a single innocence (his own) might not be opprest and overborn by a crew of mouths for the restoring of a law and doctrin falsely and unlernedly reputed new and scandalous. God... hath unexpectedly rais'd up as it were from the dead... one famous light of the first reformation to bear witnes with me Milton believed that a translation of Bucer's words would convince Parliament of the truth behind his previous tract on divorce. In the translation, he omits many sections that did not support Milton's purpose and added slight translations that connected to Milton's personal state. Notes References Miller, Leo. John Milton among the Polygamophiles. New York: Loewenthal Press, 1974. Milton, John. Complete Prose Works of John Milton Vol II ed. Don Wolfe. New Haven: Yale University Press, 1959. Patterson, Annabel. "Milton, Marriage and Divorce" in A Companion to Milton. Ed. Thomas Corns. Oxford: Blackwell Publishing, 2003. Category:Works by John Milton Category:1644 books
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Mazra, Qubadli Mazra is a village in the Qubadli Rayon of Azerbaijan. Mazra is Azeri village in Qubadli References Category:Populated places in Qubadli District
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
1884 in China Events in the year 1884 in China. Incumbents Guangxu Emperor (10th year) Regent: Empress Dowager Cixi Events Sino-French War March 6–24 - Bắc Ninh Campaign May 11 - Tientsin Accord June 23–24 - Bắc Lệ ambush August 23–26 - Battle of Fuzhou October 2–15 - Kep Campaign November 19 - Battle of Yu Oc Xinjiang province (Sinkang) established Births February 20 - Yang Sen (1884-1977) Fang Junying (1884-1923), Revolutionary Tongmenghui member Deaths Zhang Shusheng Category:Years of the 19th century in China
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Eremophila cuneifolia Eremophila cuneifolia, commonly known as pinyuru is a flowering plant in the figwort family, Scrophulariaceae and is endemic to the north-west of Western Australia. It is a spreading, sticky shrub with wedge-shaped leaves, coloured sepals and usually deep purple flowers. Description Eremophila cuneifolia is a spreading shrub growing to a height of between with hairy branches and with branches and leaves that are thickly covered with sticky, shiny resin. The leaves are arranged alternately and clustered near the ends of the stems, mostly long, about wide, wedge-shaped or heart-shaped but with a distinct small point at the tip. They are thick and stiff and the top surface of the leaves is glabrous while the lower surface is hairy. The flowers are borne singly in leaf axils on hairy stalks long. There are 5 cream-coloured to pale reddish-purple, slightly overlapping, egg-shaped sepals differing in size from each other but mostly long. The petals are long and joined at their lower end to form a tube. The petal tube is usually deep purple on the outside, sometimes paler, and white inside with purple spots. There are scattered hairs on the outside of the tube and on the lobes but the tube is filled with long, soft hairs. The 4 stamens are fully enclosed within the tube. Flowering occurs mostly from June to October and is followed by fruits which are oval to cone-shaped, shiny and long. Taxonomy and naming The species was first formally described by Friedrich Wilhelm Ludwig Kraenzlin in 1929 and the description was published in Repertorium Specierum Novarum Regni Vegetabilis Beihefte. The type specimen was collected by the explorer Charles Crossland, near Mount Hale on the upper reaches of the Murchison River, about west north west of Meekatharra. The specific epithet (cuneifolia) is from the Latin words cuneus meaning "wedge" and folium meaning "a leaf". Distribution Pinyuru occurs between Leonora, Jigalong and Onslow in the Carnarvon, Little Sandy Desert, Central Ranges, Murchison, Pilbara and Gascoyne biogeographic regions. It grows in a range of soils and habitats, often in association with other Eremophila species and is sometimes the dominant plant species. Conservation Eremophila cuneifolia is classified as "not threatened" by the Government of Western Australia Department of Parks and Wildlife. Use in horticulture This is one of the most attractive eremophilas in cultivation with its colourful sepals and deep purple flowers. It is not easy to propagate or to maintain, especially in areas like Sydney where the humidity is high. It is usually propagated by grafting onto Myoporum and needs a well-drained soil. Pinyuru is also frost sensitive and for this reason is probably best grown in a container which can be moved as the seasons change. References cuneifolia Category:Eudicots of Western Australia Category:Plants described in 1929 Category:Endemic flora of Western Australia
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
2002 Winter Paralympics medal table The 2002 Winter Paralympics medal table is a list of National Paralympic Committees (NPCs) ranked by the number of gold medals won by their athletes during the 2002 Winter Paralympics, held in Salt Lake City, Utah, United States, from March 7 to March 16, 2002. Medal table The ranking in this table is based on information provided by the International Paralympic Committee (IPC) and is consistent with IPC convention in its published medal tables. By default, the table is ordered by the number of gold medals the athletes from a nation have won (in this context, a "nation" is an entity represented by a National Paralympic Committee). The number of silver medals is taken into consideration next and then the number of bronze medals. If nations are still tied, equal ranking is given and they are listed alphabetically by IPC country code. ''To sort this table by nation, total medal count, or any other column, click on the icon next to the column title. References See also 2002 Winter Olympics medal table External links International Paralympic Committee Medal table Category:Winter Paralympics medal tables
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Monico Puentevella Monico O. Puentevella (born 2 July 1946 in Manila, Philippines) is a Filipino politician. Formerly a member of Lakas-Kampi-CMD, he was elected to three terms as a member of the House of Representatives of the Philippines, representing the Legislative district of Bacolod City. He was first elected to Congress in 2001, and was re-elected in 2004 and 2007. He was also the mayor of Bacolod from 2013-2016. He finished both his elementary and secondary education at the De La Salle Bacolod in 1962 and 1965, respectively. He obtained his Bachelor of Arts in Political Science degree from De La Salle University Manila in 1960. He furtherly pursued Professional Management Development Program study at the Asian Institute of Management in 1994. Media career Puentevella had been a contributing writer for Manila Bulletin newspaper and Sports Weekly Magazine and Program host for West Negros Radio Station in 1967. He also became part of the television world when he hosted "Monico and Friends" and "Awit ng Tanghalan" in ABS-CBN Broadcasting Corporation, and "Monico" (he also produced) in RPN TV 8. Sports career Prior to his election to Congress, Puentevella served as a Commissioner of the Philippine Sports Commission from 1996 to 2001. Puentevella served as the president of the Philippine Weightlifting Association (PWA) until 2013, when he resigned from the post upon becoming mayor of Bacolod. He then held the title president emeritus after his resignation until he was re-elected again to the post in October 2016. He succeeded Roger Dullano, who irrevocably resigned to went to the United States due to concerns related to Dullano's family. Political career 'Newks’ as he is fondly called by his contemporaries is not new to public service as he had represented the Lone District of Bacolod in Congress from 2001 to 2010. As a three-term Congressman, Mayor Monico authored thirteen National Laws including the Quarantine Act of 2004, the Cheaper Medicines Law, the Civil Aviation Authority of the Philippines Law and the Alternative Fuel Act of 2004, among others. He also helped authored and co-authored at least 586 house bills and resolutions of Local and National importance. Mayor Monico’s name is also synonymous with sports; the former De La Salle Manila stalwart was once voted Chairman of the Philippine Olympic Committee, Commissioner of the Philippine Sports Commission and the current Marketing and Television Committee Director of the Federation International de Football Association (FIFA). Helping him in his administration will be a consortium of past, present and future "servant-leaders" of the City that follows Mayor Puentevella’ s philosophy to "work as a team and succeed as one." In as such, Bacolod is assured that she is in good hands, ready and willing to upraise her name as one of the prime cities and the top tourist hub of the country. His son Claudio "Kalaw" A. Puentevella is an incumbent councilor in Bacolod. His other son Monico “Nicky” A. Puentevella, Jr. was appointed as Director of John Hay Management Corporation in 2017. Mayor of Bacolod He was elected as mayor of Bacolod City in 2013. In 2015, Puentevella was suspended for 90-days which started on November 16 and was charged with violation of the Anti-Graft and Corrupt Practices Act (Republic Act 3019) for the alleged overpriced purchase of IT packages worth P26 million using his Priority Development Assistance Funds when he was congressman from 2002 to 2007. These IT packages were for public elementary and secondary schools in Bacolod. Also charged with Puentevella were the late Victorino Tirol, former Department of Education Western Visayas director; and Merryland Publishing chairman Jessie Garcia, supplier of the IT packages.
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Puentevella denied the accusations and questioned the validity of the information filed against him, however he willfully accepted the suspension with no reported hesitation. As a result, while Puentevella served his suspension, Bacolod Vice Mayor Greg Gasataya sat as acting mayor and Councilor El Cid Familiaran as vice mayor. The Sandiganbayan’s 90-day preventive suspension of Puentevella over graft charges ended on February 14, 2016 and Puentevella was able to return to his previous post as Bacolod City Mayor. Puentevella ran for Mayoralty re-election on the 2016 Philippine local elections but lost to Evelio "Bing" Leonardia. It was reported on June 23, 2016 that the Ombudsman ordered a one-month suspension against Puentevella due to failure to immediately suspend two Bacolod City building officials. In July 2014, the Ombudsman ordered the suspension of two building officials after finding them guilty of "conduct prejudicial to the best interest of the service". Puentevella received a directive from the Ombudsman calling for the implementation of the suspension but issued a reply that the officials concerned are to be suspended in January 1, 2015 citing the unilateral move as in line with "the spirit of Christmas". The Ombudsman emphasized on its ruling suspending Puentevella that the earlier suspension order is "immediately effective and executory" and that "there is nothing vague or ambiguous with this wording". The Ombudsman also said that Puentevella has no authority to move the suspension to a different date. Puentevella said that the one-month suspension is "moot and academic" since he is already stepping down from his post as mayor eight days from the date of the report. Other involvements Monico also worked as company officer for IPC-Sugar Trading Co., Philsugar, Sugar Planters, RH Fabricators and Designs Inc., and JOSMON Management Co. Inc. He also was an employee at Philippine National Bank. In 1994, he became the President of Cebu Rolling Hills Memorial Chapels and Rolling Hills Foundation. References Category:People from Bacolod Category:1946 births Category:Living people Category:Lakas–CMD politicians Category:Members of the House of Representatives of the Philippines from Bacolod Category:Mayors of Bacolod Category:De La Salle University alumni Category:Asian Institute of Management alumni
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Waddinxveen Noord railway station Waddinxveen Noord is a railway station in northern Waddinxveen, Netherlands. The station opened on 3 June 1973 and is on the Gouda–Alphen aan den Rijn railway. The train services are operated by Nederlandse Spoorwegen. Train services The following train services call at Waddinxveen Noord: 2x per hour local service (sprinter) (Leiden -) Alphen aan den Rijn - Gouda (Operates to Leiden at peak hours only) Bus Services 175 (Rotterdam Alexander - Rotterdam Nesselande - Waddinxveen - Waddinxveen Noord) 187 (Oegstgeest - Leiden - Zoeterwoude-Rijndijk - Hazerswoude - Waddinxveen - Gouda) External links NS website Dutch Public Transport journey planner Category:Railway stations in South Holland Category:Railway stations opened in 1973
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Saint-Coutant, Deux-Sèvres Saint-Coutant is a commune in the Deux-Sèvres department in western France. See also Communes of the Deux-Sèvres department Notes Bibliography Maurice Poignat, Le Pays mellois (Poitiers: Projet Editions, 1982. ) pp. 264–267 External links INSEE Méridien Green French-English Association Category:Communes of Deux-Sèvres Category:Deux-Sèvres communes articles needing translation from French Wikipedia
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Baster (disambiguation) The Basters are an ethnic group in Namibia. Baster may also refer to: Baster, a kitchen utensil. See basting (cooking) Baster (band), French musical group People with the surname Job Baster (1711–1775), Dutch physician and naturalist Marek Baster (born 1976), Polish footballer
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Kowalówka, Podkarpackie Voivodeship Kowalówka is a village in the administrative district of Gmina Cieszanów, within Lubaczów County, Subcarpathian Voivodeship, in south-eastern Poland. It lies approximately east of Cieszanów, north of Lubaczów, and east of the regional capital Rzeszów. References Category:Villages in Lubaczów County
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Raymond Couraud Raymond Couraud (aka Captain Jack William Raymond Lee, born 12 January 1920 at Surgères, Charente-Maritime - 1977), was a French soldier and gangster, who through his World War II military exploits became a highly decorated member of the French-section of the British Army's Special Air Service. Biography Raymond Couraud was born 12 January 1920 at Surgères, Charente-Maritime, France. Little is known of his early life, or education. French Foreign Legion On 19 March 1938, Couraud joined the French Foreign Legion under his real name, but with a birth date of 12 January 1916 in Namur, Belgium. This would have made him 22 and old enough to join, whereas he was actually just over 18 and too young. After training, on 3 March 1940, he was assigned to the 5th Company of the 13th Brigade. With this unit he took part in the Battles of Narvik, for which he won the Croix de Guerre with palm, for dislodging two enemy soldiers with a grenade attack: Returning to France, Courard found a France in turmoil, in July 1940 he joined retreating French Forces in Fuveau, near Marseille. After the unit was decimated in action, while trying to reach England, Couraud was captured and imprisoned in August 1940 at Fort Saint Nicolas. Acquitted before a military tribunal in December 1940, he was released by the Vichy government. Marseille and Mary Jayne Gold On release, Couraud joined the new French economy by becoming a gangster, arranging importation, trading, distribution and export of illegal goods and people. It was through this activity that he was introduced to American socialite Mary Jayne Gold, who had chosen to relocate from Paris to the new Vichy controlled area, over returning home. In Marseilles, Gold met American art student Miriam Davenport, and American journalist and intellect Varian Fry, who had come to France on a personal mission to help members of Europe's intellectual and artistic community escape the Nazi threat. The three became the core of a volunteer group who sheltered artistic refugees, and through Couraud organized their escape through the mountains to Spain, or smuggling them aboard freighters sailing to North Africa, or ports in North or South America. Through the organisation, Couraud and Gold started a relationship. Gold helped subsidize the organisation, which is credited with the rescue of over 2,000 refugees, among whom were: sculptor Jacques Lipchitz; artist Marc Chagall; writer Hannah Arendt; Nobel Prize winner Otto Meyerhof. While Gold remained in Marseille until Autumn 1941, in April 1941 Couraud crossed the Pyrenees with the aim of reporting to the British Consulate in Barcelona. Arrested at Madrid railway station, he was interned for four months in Miranda. Meanwhile, due to criminal activity and associated smuggling of people, he was sentenced by a military court in Marseille to 10 years in prison. SOE: Jack William Raymond Lee On arrival in England on 12 October 1941, Couraud joined the Free French Forces. He was immediately assigned to the Action militaire section of the Bureau Central de Renseignements et d'Action (English: Central Bureau of Intelligence and Operations; commonly referred as just BCRA), the World War II era forerunner of the SDECE French intelligence service. Couraud was assigned to Colonel Maurice Buckmaster's section of the Special Operations Executive, which covered all of France. After specialist training, in December 1941, Couraud was commissioned as second-Lieutenant Jack William Raymond Lee. Name changes were not uncommon during the war, as various people and particularly criminals tried to hide their past while fighting what they saw as a true fight. St Nazaire Raid In January to February 1942, Couraud continued his SOE
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
training with a combination of formal courses, including parachuting; to night training on British Southcoast beaches, and reconnaissance raids on northern French beaches from Normandy to Loire-Atlantique. On 28 March, Couraud became the only French national to participate in the Lord Mountbatten-led British Commandos raid on St. Nazaire. Arranged to ensure that the Tirpitz had no base on the Atlantic Ocean, Couraud escaped, but was wounded in both legs. He recovered in Falmouth Hospital from April to July 1942. Couraud returned to operations with SOE in August 1942,with a beach raid near Cannes. In November 1942, in an operation near Narbonne, he was forced to neutralise three Vichy policeman. Escaping through the Pyrenees, he made his way to Barcelona, and then onwards to Lisbon, Portugal for return to England. 2nd SAS Regiment Because of repeated indiscretions and security breaches throughout the year 1942, Raymond Couraud is dismissed from the S.O.Es in January 1943 but recommended for transfer to a commando unit. He then joins the 62nd Commando under the command of Colonel Bill Stirling, the elder brother of SAS founder Colonel David Stirling. Assigned to Stirling's No.62 Commando, the unit later became 2nd regiment of SAS, a designated Small Scale Raiding Force. Based at the regimental HQ in Philippeville, Algeria, over the following 14 months the group undertook various raids into occupied Europe and North Africa: Ousseltia front Airports in southern Tunisia, with 21 aircraft destroyed on the ground at Metlaoui Sea born operation on the Galite Islands Sea born operation on Lampedusa Submarine operation on airfields in Sardinia SAS, French 2nd Squadron In March 1943, after a meeting General Alphonse Juin, Stirling took 50 French soldiers out of 2 SAS to form the 2 SAS, French 2nd Squadron under Couraud, under his new title of Captain Lee. The squadron consisted of Couraud as captain and commander, three British lieutenants, two French officers (Lieutenant Robert Sablet, Raillard and Aspiring), and French soldiers, with a large proportion of former Legionnaires. In May 1943, taking advantage of the Churchill Act of 1940, Couraud took British citizenship. In September 1943, in Operation Slapstick the Allied assault on Taranto, French 2nd Squadron undertook initial reconnaissance. After the landings, they destroyed a large Wehrmacht convoy in the area of Chieti, and then harassed German units as they withdrew to Sangro via the advancing front. After further raids and operations on the Italian mainland, in April 1944 the unit withdrew for recuperation in Scotland. Operation Gaff After Couraud returned to the UK, he was commanded to set up a specialist six-man assassination unit in preparation for D-Day and the Invasion of Europe, which would target high-ranking Nazi Wehrmacht staff. From March 1943, British Intelligence had been trying to find Field Marshal Erwin Rommel's headquarters in France. Part of the brief looked at creating a timetable for Rommel, to assess how easy it would be to kill him. Much as though he had been against the plan pre-D-Day, with manpower losses escalating Field Marshal Bernard Montgomery approved the plan post D-Day, even though they still didn't know where Rommel's base was. In a bizarre coincidence, on 9 June SAS Lieutenant General was heading for a meeting with a French Resistance unit. Resting in a barn, he was approached by a man who lived in La Roche-Guyon, asking for assurances that as Rommel's headquarters was in the village, that the Royal Air Force would not bomb them. On learning of the location of Rommel's HQ, Couraud and his unit were moved to a flat in London, to wait for a storm to pass and be parachuted into Orléans.
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Arriving on 18 July, the team radioed in, but found that Rommel had been severely injured the previous day after his staff car had been over turned in an attack by RAF Hawker Typhoon's, and replaced by Günther von Kluge. Couraud and his team orders regarding Rommel were aborted, and they moved toward advancing US Army lines on foot, while ambushing trains and attacking German units along their route. After an attack on the German Command at Mantes, Couraud disguised as a policeman made his way through the judicial police of Pontchartrain, and joined the American 3rd Army of General George S. Patton on 12 August. Post August 1944 After Operation Gaff, Couraud was assigned as second in command of 2 SAS under Roy Farran. Split into two teams starting from Orléans and Rennes, the teams met up near Langres, where they built an SAS field operations base. Returning to England in September 1944, with the liberation of France, Couraud left the British Army in December 1944. He returned to France to become part of the French Army General Staff. Later memorial In Varian Fry's later movie Crossroads Marseille, Couraud is referred to as "The Killer", due to his murder of the English language. References Category:1920 births Category:1977 deaths Category:People from Charente-Maritime Category:French gangsters Category:Soldiers of the French Foreign Legion Category:Free French military personnel of World War II Category:British Army personnel of World War II Category:Special Air Service officers Category:French expatriates in the United Kingdom Category:British people of French descent Category:Recipients of the Croix de Guerre 1939–1945 (France) Category:Recipients of the Military Cross Category:Recipients of the War Medal (Norway) Category:Members of the Bureau Central de Renseignements et d'Action
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Li Wei (artist) Li Wei (born in 1970, Hubei, China) is a contemporary artist from Beijing, China. His work often depicts him in apparently gravity-defying situations. Wei started off his performance series, Mirroring, and later on took off attention with his Falls series which shows the artist with his head and chest embedded into the ground. His work is a mixture of performance art and photography that creates illusions of a sometimes dangerous reality. Li Wei states that these images are not computer montages, but that he uses mirrors, metal wires, scaffolding and acrobatics. Wei's works have been published on the cover of the following magazines: Flash Art, Work, Magazin-Frankfurter Rundschau, Out of the Red, Zoom, Juliet, Contemporary, Theater Forum, NY Arts, Fine Arts Literature, Lapiz and Arte Al Limite. Life and work Li Wei grew up being a son of a farmer and had wanted to become an artist for a career. With family support he had the special privilege to study at a private arts school until he was tired of the school's traditional approach of teaching. However, Li Wei's struggle with art did not end early in his life. After Li Wei became independent he had to work numerous jobs (delivering food and house keeper) in Beijing in order for him to sustain himself; but more importantly he needed the money to supply his art works (at the time painting) so that he could continue to work towards his dream. After Li Wei kick started his career as an artist by working with paint, Li Wei realized a newfound passion and wanted to specialize in a different form of art. In 1999 he believed that "only performance art offers a chance to experience an action's message through one's own body". Since then Li Wei has been intentional about his art pieces and has desired to express his work publicly with a message behind each art piece. Li Wei has been expressing his art by capturing photos of his own body and images of himself to relay a message towards his audience. But before becoming a photographer Li Wei started as a performance artist simply because he wanted to document his performances. In an interview with AOL News Li Wei states "When I started doing this, it was 2000, I only used photography to record the procedure of my performance art. Then, after the 'Li Wei falls into ... ' series, I found this way of shooting as my signal." And since then Li Wei has already a performed for over 150 shots and many of these shots have risked his life. Li Wei himself admits that he shoots in extremely dangerous situations and he thinks the creation of his artwork isn't as a big of a challenge than being in the dangerous situation for his performance shootouts. Some of these dangerous scenarios include shots on high buildings, lakes, ice holes and even through the windshield of a war. Li Wei's rise to fame was through his performance of the “Falls” series in which he sticks himself in the group as a missile. Through this performance he wanted to relay the modern man's desire to hide away from the problems that occur daily. Li Wei's art is a difficult process. In pieces such as “Li wei Falls to the Ice Hole,” his head is planted and unseen, while his body remains rigid and upright. These photos are shocking and make the viewer puzzle over their achievement. Wei has a knack for creating catalytic art that awes and intrigues people fortunate enough to come across it. Many of Wei's
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
photos have layered meanings, demonstrating various aspects of Chinese society. In “Li Wei's Body of Art” by Julie Segraves, Wei comments on his “Falls” series, in which he demonstrates the shock of societal progression. Many of Li Wei's works are done through the usage of mirrors. The mirror would usually contain a hole in the middle for Li Wei to insert his head. With this technique Li Wei is able to create an “illusory superimpositions and fusions.” Within the Chinese culture the mirror possesses a great symbolic meaning. The mirror is an indicator of truth, allowing people to look at themselves and look at the mistakes they have made in life. Furthermore, there is a Chinese proverb that says “take history as a mirror”, which means that things that occurred throughout history will occur again and it is essential for individuals to recognize mistakes made from people in the past and react differently when similar issues come up again. Li Wei's art may seem humorous but they address wide range of topics from gender relations to politics. By creating art through mirrors, Li Wei has been able to express his personal thoughts, experiences and address some social issues that occur in our daily lives. Wei has been able to challenge the public audience's perception of 'reality' and 'truth' with his work. The actual shootout for his artwork takes about one or two hours but the preparation time is much longer. Wei could be inspired with an idea, an image in his head he wants to enact in real life, but it can take many months until he actual puts his idea into reality. Moreover, Wei hopes that he can one day shoot on the moon with a rocket. Li Wei's goal is to "make everything impossible possible." Li Wei is a forward thinking artist with fresh ideologies. In an interview with 'The Creators Project' Wei makes the interesting point that "we are all controlled by someone else. Our thoughts and actions are all controlled by an unseen force." What Wei is pointing out here is that there is an external force that dictates us and our behaviors. Furthermore, Wei mentions about the rapid changes that occurred in China in the last decade. As a citizen of Beijing, Wei fully experienced the rapid change within China's economy and the exponential growth in all types of markets in China. Wei's concern with the rapid growth was the confinement of space due to the rapid growth. People from the country started to move into the cities, into tall confined apartments. Wei feels like the rapid growth has caused a huge change and strain on people's perceptions and emotions. Due to destruction of old buildings and forced move into apartments people are forced into small confined spaces. As a result, people want to break free from these limitations, an issue that hasn't been a problem for the past decades. With this concern Wei has created art that expresses the feeling of 'breaking free', art pieces that show the hunger and will to break away from limitations and confinement. Moreover, Wei discusses his message behind his Fall series as they are not about belonging to earth, but rather it's about looking at the earth with the perspective from outer space. Wei wants his audience to look at things with a different perspective. Wei wants to draw out the fine line between reality and fantasy and point it out to the audience. And Wei intends to relay his perception of this world through his photography and performance art. Exhibitions Li Wei's artwork has been exhibited in
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Asia, Australia, Europe, and North America. Solo Exhibitions include Parc de la Villette, Paris (2012), Lucca, Italy (2011), 10 Chancery Lane Gallery, Hong Kong (2010;2006), Shanghai Tang, Hong Kong (2010), Tribeca, Madrid (2009), Michael Schultz Gallery, Beijing (2009), Mogadishni CPH gallery, Denmark (2009), Yeh Rong Jia Culture & Art Foundation, Taiwan (2008), EScape Cultural Ample Gallery, Spain (2007), PYO Gallery, Seoul (2006), Galeria Espacio Minimo, Madrid (2006), Marella Gallery, Beijing (2005), Marella Gallery, Italy (2004) Li Wei's work was included in Katonah Museum of Art, United States (2012), Beijing Times Art Museum, China (2011), Daegu Photo Biennale 2010 in Korea (2010),Palazzo Reale Museum, Milan (2009), Olympic Museum, Switzerland (2008), Criterion Gallery, Australia (2007), Toulouse Art Museum, France (2006), Ox Warehouse, Macau (2005), MMAC exhibition, Japan (2004), Prague Biennial in Czech Republic (2003), Beijing Red Square, Beijing (2002), Hong Kong Arts Center, Hong Kong (2001), Performance Art Festival, Beijing (2000) Mirror Series Li Wei began with the performance series “Mirroring” in 2000. In a mirror about a yard wide he cut a hole for his head. The mirror, propped by his hands, can be seen in many photos documenting the performances in private places but also in many public ones in China and abroad. Long shots show the observers as well as their surroundings mirrored round the performer's head. Close-ups are more striking in excluding the mirror's rim and the artist's body. Li's head seems to hover, be it in the midst of a crowd of observers, in a gully between buildings, in the sky or on glittering water. “The mirror turns concrete reality into an immaterial image by establishing new relationships between things. The process mirrors that of imagination, since we are always putting single impressions together to form an imaginary reality. Insofar as my head is stuck in the middle of this reflection one of the important functions of contemporary art becomes evident: it is unsettling – like this performance – and questions our everyday habits of perception. We see ourselves and also the surrounding reality anew.” Fall series One of his most effective series of performances has been his “Falls”, begun in 2002. Photos show the artist with his head and chest embedded in the asphalt of a street, the roof of a ruined house or the ice of a lake with his legs pointing up to the sky. “No, these images are not computer montages,” the artist tells us. Sometimes he worked with the help of props in the literal sense of the word. But for him the main thing was physical exertion, the experience – be it brief – of keeping a posture up and of feeling the absurdity of the situation through his own body. “If you picture someone falling to earth from another planet… it would really be no soft landing in the sense of a happy moment, whether the landing were in China or in another part of the world: It’s crazy what we do to one another. And this feeling of having fallen headfirst into something and of having nothing firm under the feet is familiar to everyone. One doesn’t have to fall from another planet to feel it.” Award Wei has received the following awards for his work: Macau Art Museum Overseas Communicate the Prize (Macau Art Museum, 2005) 31 Photographers with the Most Creative World (American Getty, 2006) References External links Artist's Website "The amazing pictures of an artist dangling horizontally out of a skyscraper", Daily Mail "Li Wei biography" "Artwork of LI Wei" Category:Chinese performance artists Category:Living people Category:1970 births Category:Artists from Hubei
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Radoslav Jović Radoslav Jović () is a politician and medical doctor in Serbia. He served three terms in the National Assembly of Serbia between 1997 and 2008, originally with the Serbian Renewal Movement (Srpski pokret obnove, SPO) and later with the breakaway Serbian Democratic Renewal Movement (Srpski Demokratski Pokret Obnove, SDPO). He was also the mayor of Kraljevo from 2003 to 2005. In 2014, Jović returned to the assembly as a member of the Serbian Progressive Party (Srpska napredna stranka, SNS). Early life and career Jović was born in Kraljevo, then part of the People's Republic of Serbia in the Federal People's Republic of Yugoslavia. He trained as a doctor and conducted research on malignant tumours. In his fourth year of medical school, he promoted the notion of cancer as a form of evolutionary regression on the cellular level and suggested that it could be countered through the selective use of viruses. His research was covered in the Serbian media and published in book form in 1997. When American researchers published similar findings in 2005, some suggested the possibility of a direct influence. Political career SPO representative, mayor, and SDPO representative Jović first entered the National Assembly of Serbia after the 1997 Serbian parliamentary election, in which he received the lead position on the SPO's electoral list for the Kraljevo division. The party won two mandates in the division, and he was declared elected. The Socialist Party of Serbia and its allies won the election, the SPO served in opposition. Jović served as president of the Kraljevo municipal assembly in this period. In 1998, he helped establish the annual Kraljevdan event as a celebration of the city's royalist history. He called on other Serbian towns to take in more Kosovo Serb refugees the following year, after Kraljevo's resources were challenged by a large number of refugees converging on the city. Serbia's political system was reformed for the 2000 parliamentary election, the first to be held after the downfall of Slobodan Milošević's administration. The entire country became a single electoral district, and mandates were awarded to parties instead of individual candidates, with the result that successful parties often distributed their mandates out of numerical order to candidates on their electoral lists. The SPO ran its own slate rather than joining a broad coalition called the Democratic Opposition of Serbia (DOS), which consisted of several parties that had opposed Milošević's rule. This proved to be a strategic error; the DOS won a landslide victory while the SPO did not receive a sufficient number of votes to cross the electoral threshold to win representation in the assembly. Jović, who was included in the SPO list, saw his parliamentary career come to a temporary end. Jović became the mayor of Kraljevo via direct elections in 2003, defeating a candidate of the Democratic Party of Serbia in a run-off. As mayor, he oversaw privatization proceedings for Magnohrom, a lead industrial firm in the city. He resigned in 2005 rather than defending his position in a recall referendum that had been initiated by the Serbian Radical Party; Jović argued that he did not want to defend his position at taxpayers' expense. He later challenged the legality of the recall proceedings. The SPO contested the 2003 parliamentary election in an alliance with New Serbia. Jović received the tenth position on the combined list and was selected as an SPO delegate after the list won twenty-two mandates. He took his seat when the assembly met in 2004. The SPO was part of Vojislav Koštunica's coalition government in this period, and Jović served as part of the government's parliamentary
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
majority. The SPO experienced a serious split in 2005, and several of its members, including Jović, joined the breakaway Serbian Democratic Renewal Movement. The party contested the 2007 parliamentary election as a junior partner on a coalition list led by the Democratic Party of Serbia and New Serbia; Jović received the ninety-seventh list position. The coalition won forty-seven seats, and Jović was one of two SDPO candidates to be awarded a new mandate. The party provided support to Vojislav Koštunica's government, and Jović again served with the government's parliamentary majority. The SDPO did not contest the 2008 election. SNS representative The SDPO ceased to exist in 2010. Jović later joined the Serbian Progressive Party. Serbia's electoral system was reformed in 2011, such that parliamentary mandates were awarded in numerical order to candidates on successful lists. Jović received the eighty-second position on the Progressive Party's Aleksandar Vučić — Future We Believe In list in the 2014 parliamentary election and returned to the assembly when the list won a majority with 158 out of 250 mandates. For the 2016 election, he was dropped to the 136th position the successor Aleksandar Vučić – Serbia Is Winning list. The Progressive-led alliance won a second consecutive majority with 131 mandates; Jović was not immediately re-elected but was able to take his seat once again on October 6, 2016, following the resignation of other Progressive Party members further up the list. References Category:1957 births Category:Living people Category:Members of the National Assembly of Serbia Category:Mayors of places in Serbia Category:People from Kraljevo Category:Serbian Renewal Movement politicians Category:Serbian Democratic Renewal Movement politicians Category:Serbian Progressive Party politicians
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Oenite Oenite is a mineral discovered in the Tunaberg Cu-Co-sulfide skarns, Bergslagen, Sweden, with the formula CoSbAs. References Category:minerals
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Rouleau Rouleau (French for "roll" or "roller"; plural rouleaux) can refer to: People For a list of people with Rouleau as a surname, see Rouleau (surname). Alexandre Rouleau, Canadian ice hockey player Alfred Rouleau, businessman Joseph Rouleau, opera singer Duncan Rouleau, comic book writer and artist Places Rouleau, Saskatchewan Things rouleaux, a stack of red blood cells (textiles) A decorative technique that involves creating patterns with piping, cording or bias tape. A rouleau loop uses the same cord or piping as a way of fastening buttons, most notably down the back of bridal gowns. See also Reuleaux (disambiguation)
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Epermenia davisi Epermenia davisi is a moth in the family Epermeniidae. It was described by Reinhard Gaedike in 2010. It is found in Taiwan. References Category:Moths described in 2010 Category:Epermeniidae Category:Moths of Taiwan
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Thermomycolin Thermomycolin (, thermomycolase) is an enzyme. This enzyme catalyses the following chemical reaction Rather nonspecific hydrolysis of proteins. Preferential cleavage: Ala-, Tyr-, Phe- in small molecule substrates This peptidase is isolated from the thermophilic fungus Malbranchea pulchella. References External links Category:EC 3.4.21
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
1839 English cricket season 1839 was the 53rd season of cricket in England since the foundation of Marylebone Cricket Club (MCC). Sussex County Cricket Club was formally established as the first county cricket club and replaced the ad hoc county elevens representing the traditional county of Sussex in top-class cricket. Important matches 1839 match list Events 1 March. Formation of Sussex County Cricket Club out of the Sussex Cricket Fund organisation that had been set up in 1836. The new Sussex club played its first important match against MCC at Lord’s on 10 & 11 June. Leading batsmen Ned Wenman was the leading runscorer with 332 @ 18.44 Other leading batsmen were: CG Taylor, F Pilch, J Cobbett, T Sewell, A Mynn, J Guy, GL Langdon Leading bowlers James Cobbett was the leading wicket-taker with 85 Other leading bowlers were: FW Lillywhite, J Bayley, E Sayres, A Mynn, WR Hillyer, GB Lee, S Redgate, J Dean, CG Taylor, JH Kirwan References Annual reviews Arthur Haygarth, Scores & Biographies, Volume 2 (1827-1840), Lillywhite, 1862 External links CricketArchive – season summaries Category:1839 in English cricket Category:English cricket seasons from 1816 to 1863
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Romania national under-20 rugby union team The Romania national under-20 rugby union team is Romania's junior national team in rugby union. They have represented Romania in the IRB Junior World Rugby Trophy now known as World Rugby Under 20 Trophy, but not yet in the World Rugby Under 20 Championship. They participated in the first three Junior World Trophies, winning it in 2009, and in 2018 they will be the hosts of the World Rugby Under 20 Trophy for the first time. Results 2008 IRB Junior World Rugby Trophy Pool 28 – 26 46 – 7 3 - 14 3rd place play-off 10 – 34 2009 IRB Junior World Rugby Trophy Pool 50 - 17 65 - 14 26 – 20 Final 25 - 13 2010 IRB Junior World Rugby Trophy Pool 15 - 12 7 - 30 48 – 12 3rd place play-off 20 – 23 2018 World Rugby Under 20 Trophy Pool 55 - 26 31 - 17 33 – 56 7th place play-off 14 – 71 Former squad Current squad Management Ovidiu Carcei - Manager Răzvan Popovici - Head Coach Adrian Costache - Fowards Coach Valentin Calafeteanu - Skills Coach Lynn Howells - World Rugby Technical Consultant Petre Neacșu - Physical Preparator Ilie Vlad - Medic Dinu Horea - Kinetotherapist See also Rugby union in Romania External links Official website Under20 Category:European national under-20 rugby union teams
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Alfred Teall Alfred Teall was a priest who worked as a missionary for the Anglican Church in Melanesia. Teall was educated at Dorchester Missionary College and ordained in 1921. He was on Ambae Island from 1920 to 1928; and Banks Island from 1928 to 1935. He was Archdeacon of Southern Melanesia from 1935 to 1959. References Category:Anglican missionaries Category:20th-century Anglican priests Category:Alumni of Dorchester Missionary College Category:Archdeacons of Southern Melanesia
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
New Forwarding Company JSC New Forwarding Company or NFC (), is one of the largest independent railway carriers in Russia. It was established in 2003. NFC transports products of oil and mining industries, iron-and-steel works, coal, construction materials, and other goods. CEO of the company is Valeriy Shpakov. Activity NFC has its own fleet of 21,000 freight wagons (including tanks, open wagons, flat-cars), locomotives, all day round control center, and wagon repair service. NFC transportation geography includes the entire territory of Russia, Lithuania, Latvia, Estonia, Ukraine, Belarus and China. NFC has 10 operational branch offices in Russian Federation. References External links JSC New forwarding company Category:Railway companies of Russia Category:Companies based in Moscow
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Alsancak Gar (Tram İzmir) Alsancak Gar is a light-rail station on the Konak Tram of the Tram İzmir system in İzmir, Turkey. It is located along Liman Avenue, next to Alsancak station. Transfer to İZBAN commuter rail service is available at Alsancak station. Alsancak Gar station opened on 24 March 2018. Connections ESHOT operates city bus service on Liman Avenue. Nearby places of interest Alsancak railway station Yzb. Şerafettin Bey Street - Plenty of Greek houses from the Ottoman Empire. İzmir Towngas Factory - A former towngas plant, turned into a museum and open air theater. Pictures References Category:Railway stations opened in 2018 Category:2018 establishments in Turkey Category:Konak District Category:Tram transport in İzmir
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
Joannes Molanus Joannes Molanus (1533–1585), often cited simply as Molanus, is the Latinized name of Jan Vermeulen or Van der Meulen, an influential Counter Reformation Catholic theologian of Louvain University, where he was Professor of Theology, and Rector from 1578. Born at Lille (a city in the County of Flanders, then under Habsburg rule), he was a priest and canon of St. Peter's Church, Leuven, where he died. He wrote numerous books, several only published posthumously. He is best known for his De Picturis et Imaginibus Sacris, pro vero earum usu contra abusus ("Treatise on Sacred Images"). This was published in 1570, four years after the Iconoclastic Fury had swept through the Low Countries, and it defended the production and use of devotional images, but enforcing the restrictions of the Council of Trent, as he interpreted them, in a brutally polemical fashion, which was very influential. Five further, enlarged, editions of this appeared between 1594 and 1771, and a modern French translation was published in 1996. He was also lead editor of an edition of the works of Saint Augustine (Antwerp, Plantin Press, 1566–1577), and wrote a manuscript history of Louvain that was printed in two volumes in 1861, edited by P. F. X. de Ram. Life Molanus was born in Lille, in Walloon Flanders, in 1533, the son of Hendrik Vermeulen and Anna Peters. His father was from Holland and his mother from Brabant. He matriculated at Louvain University on 27 February 1554, graduating in the Liberal Arts in 1558 and as Doctor of Theology in 1570. He sat on the committee of theologians overseeing Lucas Brugensis's revision of the Leuven Vulgate, published in 1574. He became a canon of St. Peter's Church, Leuven, and a professor of Theology, serving both as dean of the Faculty of Theology and as rector of the university. In 1579 he was appointed president of King's College. Molanus died in Louvain on 18 September 1585, having made bequests in favour of the college over which he had presided. Molanus and art Molanus is today mainly remembered by art historians for being one of the first authorities to turn the Council of Trent's short and inexplicit decrees on sacred images (1563) into minutely detailed instructions for artists, which were then widely enforced in Catholic countries mostly. His views on the older, originally Byzantine, depiction of the Nativity of Jesus in art are typical: The Virgin is shown pale with pains, the midwives prepare a small (narcotic) drought for the childbirth. Why this? Is it because the Virgin Mary would have held back from any pain of childbirth, when in fact she brought forth her divine son without pain? And what pertains to the midwives who are mentioned in the apocryphal Book of the Infancy? Jerome says: There was no midwife! No obtrusiveness of women intervened! She, the Virgin, was both mother and midwife! I saw in not a few places the picture of the blessed Virgin lying on a bed, depicting childbirth, and she was suffering pains from this birth, but that is not true. How stupid! Those artists ought to be laughed at who paint Mary in the very act of childbirth pains, accompanied with pain, midwife, bed, little knives (to cut the umbilical cord), with hot compresses, and many other appurtenances. . . . Rather, those pictures should be promoted which show the birth of Christ in which the Blessed Virgin Mary with arms folded and on bended knee before her little son, as though he was just now brought forth into the light. He objected on similar grounds to the Death
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
of the Virgin, her swooning at the foot of the Cross, and her being shown supplicating Christ for mankind in Last Judgement scenes. She would, he said, in fact be sitting alongside Christ in stern judgement: Many painters show Mary and John the Baptist kneeling beside Our lord at the Last Judgment...But we may not think that at that day the Virgin Mary will kneel for us before the Judge, baring her breast to intercede for sinners. Nor may we think that John the Baptist will fall upon his knees to beg mercy for mankind in the way the painters show. Rather, the blessed Virgin and St. John shall sit beside the supreme Judge as assessors. The mercy which is extended now will have no place then. There will only be strict justice at that day. Also condemned were Saint Christopher as a giant carrying Christ (and as the protector of travellers), Saint George and the Dragon, the Holy Kinship, the unicorn hunt in the Hortus Conclusus and many other depictions not verifiable from reliable sources – the Golden Legend was, he said, in fact of "lead". Nudity, even of the infant Jesus, was to be avoided as much as possible, and drapes must cover the genitals where it could not be. Whilst condemning older depictions without a scriptural basis, he was not hesitant in creating new ones based on his own interpretation. Saint Joseph should not be shown as the old, semi-comical figure of the Middle Ages, but as young, vigorous and firmly in control of the Holy Family. Mary Magdalene should not be shown as an over-dressed prostitute, and generally dress should be simple. He opposed showing the Swooning Virgin Mary in scenes of Christ carrying the Cross or the Crucifixion itself. Publications Usuardi martyrologium (Leuven, 1568) Available on Google Books De Picturis et Imaginibus Sacris (Leuven, 1570) Available on Google Books Reworked as the posthumously published Historia Sanctarum Imaginum (Leuven, 1594) Available on Google Books Indiculus Sanctorum Belgii (Leuven, 1573) Available on Google Books De fide haereticis servanda libri tres (Cologne, 1584) Available on Google Books Theologiae Practicae Compendium (Cologne, 1585) Available on Google Books Liber de piis testamentis (Cologne, 1585) Available on Google Books De Canonicis libri tres (Cologne, 1587) Available on Google Books Militia sacra ducum et principum Brabantiae (Antwerp, 1592), edited by Henricus van Cuyck for posthumous publication; available on Google Books Natales sanctorum Belgii (Antwerp, 1595), edited by Henricus van Cuyck for posthumous publication; available on Google Books Notes References Blunt, Anthony, Artistic Theory in Italy, 1450-1660, chapter VIII, especially pp. 107–128, 1940 (refs to 1985 edn), OUP, Decavele, Johan, "Molanus, Johannes", Grove Art Online, Oxford Art Online, Oxford University Press. Web. 4 Apr. 2017. subscription required. Mâle, Émile, L'Art religieux après le Concile de Trente, étude sur l'iconographie de la fin du XVIe, du XVIIe et du XVIIIe siècles en Italie, en France, en Espagne et en Flandre (1932) Penny, Nicholas, National Gallery Catalogues (new series): The Sixteenth Century Italian Paintings, Volume I, 2004, National Gallery Publications Ltd, Category:1533 births Category:1585 deaths Category:People from Lille Category:Flemish theologians Category:Art critics Category:University of Leuven faculty Category:Flemish Roman Catholic theologians
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.
Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.
Here is the passage you will be working with:
<passage>
2015 Hartford mayoral election The 2015 Hartford mayoral election was held on November 3, 2015. The election was won by Democrat Luke Bronin. Primaries Primaries were held September 15. Democratic primary Ahead of the primary, Luke Bronin had won the endorsement of the Democratic primary. He defeated incumbent mayor Pedro Segarra in the primary. Bronin outspent Segarra by a margin of 3 to 1. Bronin had raised $800,000 in his primary campaign. Turnout in the Democratic primary was approximately 26%. Republican nomination Theodore T. Cannon won the Republican nomination. General election Hartford is a highly Democratic city, therefore Bronin was anticipated to win the general election. References 2015 Hartford Hartford Category:21st century in Hartford, Connecticut
</passage>
Follow these steps to complete the task:
1. Inserting retrieval tokens:
a. Read through the text carefully.
b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.
c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).
d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).
e. Key Factual information should not intersect with Prefix Intent.
2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):
a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).
b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).
c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.
3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.
a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them.
b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>)
c. Ensure that the very first PI in the passage is preceded by <|start-PI|>.
d. Double-check that every PI and KFI is properly enclosed in these tags.
4. Describing ideal retrieved information:
a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus.
b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent.
c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)
5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.
6. Output format:
For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.
- Prefix_intent: [Write the Prefix Intent here]
- Prefix_reformulated: [Write the Prefix Intent reformulated]
- Key_Factual_Information: [Write the Key Factual Information here]
- Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]
- Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]
- Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]
7. Final step:
After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions
Here's an example of the expected output format:
Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.
```json
{ "results" : [
{
"Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to",
"Prefix_reformulated": "What is the height of Girmal Falls?",
"Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat",
"Verification": "The KFI directly provides the height information requested in the PI.",
"Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.",
"Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement."
},
{
"Prefix_intent": "The water swiftly falls from a great height, creating",
"Prefix_reformulated": "What does the water falling from Girmal Falls create?",
"Key_Factual_Information": "a fog like condition that's eye catching",
"Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.",
"Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.",
"Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description."
},
{
"Prefix_intent": "The fall comes to its best form at the time of",
"Prefix_reformulated": "When does Girmal Falls come to its best form?",
"Key_Factual_Information": "monsoon and provides an immensely striking appearance",
"Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.",
"Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.",
"Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed."
}],
"Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler."
}
```
8. Final check:
a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged.
b. Verify that the first PI in the passage starts with <|start-PI|>.
c. Make sure there are no missing or misplaced tags.
Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File. |