input
stringlengths 3.23k
13.3k
| output
sequencelengths 1
3
| id
stringlengths 40
40
|
|---|---|---|
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Then we extract these features from two sources of context texts, specifically the title of the news article that the comment was posted for and the screen name of the user who posted the comment.
Output:
| [
"What context do they use?"
] | task461-22fce2273c2d46aa8f891e0eec79cae8 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Since, we there was no publicly available standard Nepali NER dataset and did not receive any dataset from the previous researchers, we had to create our own dataset. This dataset contains the sentences collected from daily newspaper of the year 2015-2016.
Output:
| [
"What is the source of their dataset?"
] | task461-b766fb3e87274af28a3772bac621afc7 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Next, we analyze why they do not perform well in this task and with this data:
[noitemsep,leftmargin=*]
Noisy data. A major challenge of this task is the discrepancy between annotations due to subjective judgement. Although this affects also detection using only text, its repercussion is bigger in more complex tasks, such as detection using images or multimodal detection.
Complexity and diversity of multimodal relations. Hate speech multimodal publications employ a lot of background knowledge which makes the relations between visual and textual elements they use very complex and diverse, and therefore difficult to learn by a neural network.
Small set of multimodal examples. Fig. FIGREF5 shows some of the challenging multimodal hate examples that we aimed to detect. But although we have collected a big dataset of $150K$ tweets, the subset of multimodal hate there is still too small to learn the complex multimodal relations needed to identify multimodal hate.
Output:
| [
"What is author's opinion on why current multimodal models cannot outperform models analyzing only text?"
] | task461-9d838767dfe7485e878341c9b0556943 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Thus, to limit the number of different inputs to the classifier, we wish to reduce the number of distinct word recognition outputs that an attacker can induce, not just the number of words on which the model is “fooled”. We denote this property of a model as its sensitivity.
Output:
| [
"What does the \"sensitivity\" quantity denote?"
] | task461-75cadb826f9b44b2a953e6f33c4417e1 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The joke corpus in this dataset contains thousands of unique jokes of different categories (sci-fi, sports, etc) and types (puns, limerick, etc).
Output:
| [
"Where did the real production data come from?"
] | task461-0c5fbe3f88e74cb39ca491ba05c9b0db |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In this subsection, we see the influence of each component of a model on performance by removing or replacing its components. the SNLI dataset is used for experiments, and the best performing configuration is used as a baseline for modifications. We consider the following variants: (i) models that use plain stacked LSTMs, (ii) models with different INLINEFORM0 , (iii) models without INLINEFORM1 , and (iv) models that integrate lower contexts via peephole connections.
Output:
| [
"What were the baselines?"
] | task461-9d0c245de393412fae504b95e1e0c92d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We adapted BERTNLU from ConvLab-2. We implemented a rule-based model (RuleDST) and adapted TRADE (Transferable Dialogue State Generator) BIBREF19 in this experiment. We adapted a vanilla policy trained in a supervised fashion from ConvLab-2 (SL policy).
Output:
| [
"What are the benchmark models?"
] | task461-57ec57dd34014e93b7bdc31eac3a5e1b |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In order to decode spatial connections between the electrodes from the channel covariance matrix, we use a CNN BIBREF19 , in particular a four-layered 2D CNN stacking two convolutional and two fully connected hidden layers. The INLINEFORM0 feature map at a given CNN layer with input INLINEFORM1 , weight matrix INLINEFORM2 and bias INLINEFORM3 is obtained as: INLINEFORM4 . At this first level of hierarchy, the network is trained with the corresponding labels as target outputs, optimizing a cross-entropy cost function. In parallel, we apply a four-layered recurrent neural network on the channel covariance matrices to explore the hidden temporal features of the electrodes. Namely, we exploit an LSTM BIBREF20 consisting of two fully connected hidden layers, stacked with two LSTM layers and trained in a similar manner as CNN.
Output:
| [
"How was the spatial aspect of the EEG signal computed?"
] | task461-69e7e572de3b4a8c871134e667ae65da |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Our dataset is constructed via distant supervision from Twitter.
Output:
| [
"Where did they get the data for this project?"
] | task461-da1225742ab04332a19a626c208ce840 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We would like to thank the organizers of the WASSA-2017 Shared Task on Emotion Intensity, for providing the data, the guidelines and timely support.
Output:
| [
"what dataset was used?"
] | task461-ed80ea0b358d44359a8750f7d623a10d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We compare classification and regression approaches and show that classification produces better results than regression but the quality of the results depends on the approach followed to annotate the data labels.
Output:
| [
"Did classification models perform better than previous regression one?"
] | task461-6557468becce4c7790ae90e7f0c54398 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We pair 11'248 standard German written words with their phonetical representations in six different Swiss dialects: Zürich, St. Gallen, Basel, Bern, Visp, and Stans (Figure FIGREF1).
Output:
| [
"How many words are coded in the dictionary?"
] | task461-c6cb3349c5a54b86b979e415e874c63d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We carried out a reliability study for the proposed scheme using two pairs of expert annotators, P1 and P2.
Output:
| [
"do they use a crowdsourcing platform?"
] | task461-dce8a705bd28448f8c97ffc281029861 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Kneser–Ney smoothing In particular, we compare Kneser–Ney smoothing, widely accepted as the state of the art prior to NLMs, to the best NLMs today.
Output:
| [
"what classic language models are mentioned in the paper?"
] | task461-75260dcc6b1e43d2b70a28b3027ea6de |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Using the fine-grained gating mechanism conditioned on the lexical features, we can accurately control the information flows between word-level and char-level. Intuitively, the formulation is as follows: INLINEFORM0
where INLINEFORM0 is the element-wise multiplication operator. when the gate has high value, more information flows from the word-level representation; otherwise, char-level will take the dominating place. It is practical in real scenarios. For example, for unfamiliar noun entities, the gates tend to bias towards char-level representation in order to care richer morphological structure.
Output:
| [
"How does the gatint mechanism combine word and character information?"
] | task461-d456f0bc378f47acbe52fcc9ac8bbc63 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: To train our model, we generated a dataset of 20,000 demonstrated 7 DOF trajectories (6 robot joints and 1 gripper dimension) in our simulated environment together with a sentence generator capable of creating natural task descriptions for each scenario. In order to create the language generator, we conducted an human-subject study to collect sentence templates of a placement task as well as common words and synonyms for each of the used features. By utilising these data, we are able to generate over 180,000 unique sentences, depending on the generated scenario. To test our model, we generated 500 new scenario testing each of the three features to identify the correct target among other bowls.
Output:
| [
"Does proposed end-to-end approach learn in reinforcement or supervised learning manner?"
] | task461-dc1c423fec60408586609ac81e2f627b |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The following different sense inventories have been used during the evaluation: Watlink, a word sense network constructed automatically. It uses the synsets induced in an unsupervised way by the Watset[CWnolog, MCL] method BIBREF2 and the semantic relations from such dictionaries as Wiktionary referred as Joint INLINEFORM0 Exp INLINEFORM1 SWN in Ustalov:17:dialogue.
Output:
| [
"Which corpus of synsets are used?"
] | task461-5859c44b01dc4824927c3884a093e6be |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The main contribution of our work is thus a new parallel data-to-text NLG corpus that (1) is more conversational, rather than information seeking or question answering, and thus more suitable for an open-domain dialogue system, (2) represents a new, unexplored domain which, however, has excellent potential for application in conversational agents, and (3) has high-quality, manually cleaned human-produced utterances.
Output:
| [
"How the authors made sure that corpus is clean despite being crowdsourced?"
] | task461-5908f99c296d400fbf8de0ace0768e3a |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: To account for sequences of words and characters that might carry useful information, we extracted word unigrams, bigrams, and trigrams as features.
Output:
| [
"Which features do they use to model Twitter messages?"
] | task461-e6af755d0c78493a8f2b877c0a61527c |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We find that dynamic communities, such as Seahawks or starcraft, have substantially higher rates of monthly user retention than more stable communities (Spearman's INLINEFORM0 = 0.70, INLINEFORM1 0.001, computed with community points averaged over months; Figure FIGREF11 .A, left). Similarly, more distinctive communities, like Cooking and Naruto, exhibit moderately higher monthly retention rates than more generic communities (Spearman's INLINEFORM2 = 0.33, INLINEFORM3 0.001; Figure FIGREF11 .A, right). As with monthly retention, we find a strong positive relationship between a community's dynamicity and the average number of months that a user will stay in that community (Spearman's INLINEFORM0 = 0.41, INLINEFORM1 0.001, computed over all community points; Figure FIGREF11 .B, left). This verifies that the short-term trend observed for monthly retention translates into longer-term engagement and suggests that long-term user retention might be strongly driven by the extent to which a community continually provides novel content. Interestingly, there is no significant relationship between distinctiveness and long-term engagement (Spearman's INLINEFORM2 = 0.03, INLINEFORM3 0.77; Figure FIGREF11 .B, right). Thus, while highly distinctive communities like RandomActsOfMakeup may generate focused commitment from users over a short period of time, such communities are unlikely to retain long-term users unless they also have sufficiently dynamic content.
Output:
| [
"How do the various social phenomena examined manifest in different types of communities?"
] | task461-59c5b45c9447421d9ff3f0b1f80474cd |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: To tackle this issue we applying a back-translation method BIBREF13, where we translate indirect and physical harassment tweets of the train set from english to german, french and greek. After that, we translate them back to english in order to achieve data augmentation.
Output:
| [
"What language(s) is/are represented in the dataset?"
] | task461-87d181d7101344abb2bebd5a3930c521 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Amazon Reviews Dataset BIBREF24 is a large dataset with millions of reviews from different product categories. For our experiments, we consider a subset of 20000 reviews from the domains Cell Phones and Accessories(C), Clothing and Shoes(S), Home and Kitchen(H) and Tools and Home Improvement(T). Out of 20000 reviews, 10000 are positive and 10000 are negative. We use 12800 reviews for training, 3200 reviews for validation and 4000 reviews for testing from each domain.
Output:
| [
"For the purposes of this paper, how is something determined to be domain specific knowledge?"
] | task461-7dfe3f6551df4ff4bd7e765907414753 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: BioASQ organizers provide the training and testing data. The training data consists of questions, gold standard documents, snippets, concepts, and ideal answers (which we did not use in this paper, but we used last year BIBREF2). The test data is split between phases A and B. The Phase A dataset consists of the questions, unique ids, question types. The Phase B dataset consists of the questions, golden standard documents, snippets, unique ids and question types.
Output:
| [
"What dataset did they use?"
] | task461-3a2fe7695f60426b8e2ea7641bfb4db2 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We compare with three baseline methods. $({1})$ SC-LSTM BIBREF3 is a canonical model and a strong baseline that uses an additional dialog act vector and a reading gate to guide the utterance generation. $({2})$ GPT-2 BIBREF6 is used to directly fine-tune on the domain-specific labels, without pre-training on the large-scale corpus of (dialog act, response) pairs. $({3})$ HDSA BIBREF7 is a state-of-the-art model on MultiWOZ. It leverages dialog act structures to enable transfer in the multi-domain setting, showing superior performance than SC-LSTM.
Output:
| [
"What existing methods is SC-GPT compared to?"
] | task461-f1e8596165014d69b63b81c8619cc182 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: For the two single sentence tasks—the syntax-oriented CoLA task and the SST sentiment task—we find somewhat deteriorated performance.
Output:
| [
"Does the additional training on supervised tasks hurt performance in some tasks?"
] | task461-f9d7763fcc744cbb81b7b783268451c9 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The summaries generated by our ILP-based approach were compared with the corresponding manual summaries using the ROUGE BIBREF22 unigram score.
Output:
| [
"What evaluation metrics were used for the summarization task?"
] | task461-b0937aade18c4cd88355fdd3ee11adc6 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We use the BLEU BIBREF30 metric on the validation set for the VQG model training. BLEU is a measure of similitude between generated and target sequences of words, widely used in natural language processing. It assumes that valid generated responses have significant word overlap with the ground truth responses. We use it because in this case we have five different references for each of the generated questions. We obtain a BLEU score of 2.07.
Our chatbot model instead, only have one reference ground truth in training when generating a sequence of words. We considered that it was not a good metric to apply as in some occasions responses have the same meaning, but do not share any words in common. Thus, we save several models with different hyperparameters and at different number of training iterations and compare them using human evaluation, to chose the model that performs better in a conversation.
Output:
| [
"How is performance of this system measured?"
] | task461-5e8a2ab08d1e49b29e012fcb9952ff3f |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We propose a new unsupervised model, Sent2Vec, for learning universal sentence embeddings Formally, we learn a source (or context) embedding INLINEFORM0 and target embedding INLINEFORM1 for each word INLINEFORM2 in the vocabulary, with embedding dimension INLINEFORM3 and INLINEFORM4 as in ( EQREF6 ). The sentence embedding is defined as the average of the source word embeddings of its constituent words, as in ( EQREF8 ). We augment this model furthermore by also learning source embeddings for not only unigrams but also n-grams present in each sentence, and averaging the n-gram embeddings along with the words, i.e., the sentence embedding INLINEFORM5 for INLINEFORM6 is modeled as DISPLAYFORM0
Output:
| [
"How do the n-gram features incorporate compositionality?"
] | task461-749c226816004a26aafd58eb12024e9e |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Vanilla ST baseline: The vanilla ST BIBREF9 has only a speech encoder and a decoder. It is trained from scratch on the ST-TED corpus.
Pre-training baselines: We conduct three pre-training baseline experiments: 1) encoder pre-training, in which the ST encoder is initialized from an ASR model; 2) decoder pre-training, in which the ST decoder is initialized from an MT model; and 3) encoder-decoder pre-training, where both the encoder and decoder are pre-trained. The ASR model has the same architecture with vanilla ST model, trained on the mixture of ST-TED and TED-LIUM2 corpus. The MT model has a text encoder and decoder with the same architecture of which in TCEN. It is first trained on WMT data (out-of-domain) and then fine-tuned on in-domain data.
Multi-task baselines: We also conduct three multi-task baseline experiments including one-to-many setting, many-to-one setting, and many-to-many setting. In the first two settings, we train the model with $\alpha _{st}=0.75$ while $\alpha _{asr}=0.25$ or $\alpha _{mt}=0.25$. For many-to-many setting, we use $\alpha _{st}=0.6, \alpha _{asr}=0.2$ and $\alpha _{mt}=0.2$.. For MT task, we use only in-domain data.
Many-to-many+pre-training: We train a many-to-many multi-task model where the encoders and decoders are derived from pre-trained ASR and MT models. Triangle+pre-train: BIBREF18 DBLP:conf/naacl/AnastasopoulosC18 proposed a triangle multi-task strategy for speech translation. Their model solves the subnet waste issue by concatenating an ST decoder to an ASR encoder-decoder model. Notably, their ST decoder can consume representations from the speech encoder as well as the ASR decoder. For a fair comparison, the speech encoder and the ASR decoder are initialized from the pre-trained ASR model. The Triangle model is fine-tuned under their multi-task manner.
Output:
| [
"What are the baselines?"
] | task461-40d71e60a19245f9a016c9b7bfea90aa |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We extract 5 Surface and Lexical features, namely sequence length in number of tokens, average word length, type-token ratio, and lexical to tokens ratio (ratio of adjectives, verbs, nouns, and adverbs to tokens). We extracted two features that use a learned representation: Firstly, we get a sentence embedding feature that is built by averaging the word embeddings of an input sentence. Secondly, we extract a fastText representation using the fastText library with the same parameters as reported in Joulin et al. joulin2016bag.
Output:
| [
"Do they differentiate insights where they are dealing with learned or engineered representations?"
] | task461-ef06cf6db36b462eaedc4e51608f6531 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Tweets related to Forex, specifically to EUR and USD, were acquired through the Twitter search API with the following query: “EURUSD”, “USDEUR”, “EUR”, or “USD”. In the period of three years (January 2014 to December 2016) almost 15 million tweets were collected. A subset of them (44,000 tweets) was manually labeled by knowledgeable students of finance.
Output:
| [
"How many tweets were manually labelled? "
] | task461-da506fff255e4ec3a9285d2f1d14effc |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: To evaluate our model capability with processing the Chinese language, we also tested the performance of LCF-ATEPC on four Chinese comment datasets BIBREF35, BIBREF36, BIBREF29 (Car, Phone, Notebook, Camera).
Output:
| [
"In what four Chinese review datasets does LCF-ATEPC achieves state of the art?"
] | task461-8aa0252916c84eafa4120430c8bdeb89 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Since labels are unknown on INLINEFORM1 , EGL computes the expectation of the gradient norm over all possible labelings. BIBREF3 interprets EGL as “expected model change”. In the following section, we formalize the intuition for EGL and show that it follows naturally from reducing the variance of an estimator. Eq. ( EQREF7 ) indicates that to reduce INLINEFORM0 on test data, we need to minimize the expected variance INLINEFORM1 over the test set. A practical issue is that we do not know INLINEFORM0 in advance. We could instead substitute an estimate INLINEFORM1 from a pre-trained model, where it is reasonable to assume the INLINEFORM2 to be close to the true INLINEFORM3 . The batch selection then works by taking the samples that have largest gradient norms, DISPLAYFORM0
For RNNs, the gradients for each potential label can be obtained by back-propagation. Another practical issue is that EGL marginalizes over all possible labelings, but in speech recognition, the number of labelings scales exponentially in the number of timesteps. Therefore, we only marginalize over the INLINEFORM0 most probable labelings. They are obtained by beam search decoding, as in BIBREF7 . The EGL method in BIBREF3 is almost the same as Eq. ( EQREF8 ), except the gradient's norm is not squared in BIBREF3 .
Output:
| [
"How do they calculate variance from the model outputs?"
] | task461-40be583434e7482a9214ef91c9c144b5 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Our baseline is a three-layer LSTM-LM with 1,150 hidden units at internal layers trained with the standard cross-entropy loss.
Output:
| [
"What neural language models are explored?"
] | task461-a0a6fb1a32584a8bb5c5ae48be795355 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In order to find the object type, SimplerVoice, first, builds an ontology-based knowledge tree. We propose to use 2 methods to generate the suitable verbs for the target object: heuristics-based, and n-grams model.
Output:
| [
"Which model do they use to generate key messages?"
] | task461-684392bfd1564e94a6bd6b6ff957cc1a |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: For French speech technologies, four corpora containing radio and TV broadcast are the most widely used: ESTER1 BIBREF13, ESTER2 BIBREF14, ETAPE BIBREF15 and REPERE BIBREF16.
Output:
| [
"What are four major corpora of French broadcast?"
] | task461-9d1456e2402042e8abf54437267c6f99 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: For both datasets, we follow the evaluation metrics used in the original evaluation tasks BIBREF13 . For DBQA, P@1 (Precision@1), MAP (Mean Average Precision) and MRR (Mean Reciprocal Rank) are adopted. For KBRE, since only one golden candidate is labeled for each question, only P@1 and MRR are used.
Output:
| [
"Which metrics do they use to evaluate matching?"
] | task461-fc4044a4897b4528b4f2533506557ee6 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: For English all-words WSD, we train our WSD model on SemCor BIBREF24, and test it on Senseval-2 (SE2), Senseval-3 (SE3), SemEval 2013 task 12 (SE13), and SemEval 2015 task 13 (SE15). We use OntoNotes Release 5.0, which contains a number of annotations including word senses for Chinese.
Output:
| [
"What datasets are used for testing?"
] | task461-3906794b7bf14f84a68d80e432dcc6b2 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Metrics. We use tolerance accuracy BIBREF16, which measures how far away the predicted span is from the gold standard span, as a metric. The rationale behind the metric is that, in practice, it suffices to recommend a rough span which contains the answer – a difference of a few seconds would not matter much to the user. Metrics. We used accuracy and MRR (Mean Reciprocal Ranking) as metrics. Metrics. To evaluate our pipeline approach we use overall accuracy after filtering and accuracy given that the segment is in the top 10 videos.
Output:
| [
"What evaluation metrics were used in the experiment?"
] | task461-566e116a8bdf432a8fc86425ce69b382 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: TextCat is the most well-known Perl implementation of the out-of-place method, it lists models for 76 languages in its off-the-shelf configuration; the program is not actively maintained. TextCat is the most well-known Perl implementation of the out-of-place method, it lists models for 76 languages in its off-the-shelf configuration; the program is not actively maintained. TextCat is not the only example of an off-the-shelf implementation of the out-of-place method: other implementations include libtextcat with 76 language models, JTCL with 15 languages, and mguesser with 104 models for different language-encoding pairs. The main issue addressed by later implementations is classification speed: TextCat is implemented in Perl and is not optimized for speed, whereas implementations such as libtextcat and mguesser have been specifically written to be fast and efficient. whatlang-rs uses an algorithm based on character trigrams and refers the user to the BIBREF7 article. It comes pre-trained with 83 languages.
is the language identifier embedded in the Google Chrome web browser. It uses a NB classifier, and script-specific classification strategies. assumes that all the input is in UTF-8, and assigns the responsibility of encoding detection and transcoding to the user. uses Unicode information to determine the script of the input. also implements a number of pre-processing heuristics to help boost performance on its target domain (web pages), such as stripping character sequences like .jpg. The standard implementation supports 83 languages, and an extended model is also available, that supports 160 languages.
is a Java library that implements a language identifier based on a NB classifier trained over character . The software comes with pre-trained models for 53 languages, using data from Wikipedia. makes use of a range of normalization heuristics to improve the performance on particular languages, including: (1) clustering of Chinese/Japanese/Korean characters to reduce sparseness; (2) removal of “language-independent” characters, and other text normalization; and (3) normalization of Arabic characters.
is a Python implementation of the method described by BIBREF150 , which exploits training data for the same language across multiple different sources of text to identify sequences of characters that are strongly predictive of a given language, regardless of the source of the text. This feature set is combined with a NB classifier, and is distributed with a pre-trained model for 97 languages prepared using data from 5 different text sources. BIBREF151 provide an empirical comparison of to , and and find that it compares favorably both in terms of accuracy and classification speed. There are also implementations of the classifier component (but not the training portion) of in Java, C, and JavaScript.
BIBREF153 uses a vector-space model with per-feature weighting on character sequences. One particular feature of is that it uses discriminative training in selecting features, i.e. it specifically makes use of features that are strong evidence against a particular language, which is generally not captured by NB models. Another feature of is that it uses inter-string smoothing to exploit sentence-level locality in making language predictions, under the assumption that adjacent sentences are likely to be in the same language. BIBREF153 reports that this substantially improves the accuracy of the identifier. Another distinguishing feature of is that it comes pre-trained with data for 1400 languages, which is the highest number by a large margin of any off-the-shelf system.
whatthelang is a recent language identifier written in Python, which utilizes the FastText NN-based text classification algorithm. It supports 176 languages.
implements an off-the-shelf classifier trained using Wikipedia data, covering 122 languages. Although not described as such, the actual classification algorithm used is a linear model, and is thus closely related to both NB and a cosine-based vector space model.
In addition to the above-mentioned general-purpose language identifiers, there have also been efforts to produce pre-trained language identifiers targeted specifically at Twitter messages. is a Twitter-specific tool with built-in models for 19 languages. It uses a document representation based on tries BIBREF401 . The algorithm is a LR classifier using all possible substrings of the data, which is important to maximize the available information from the relatively short Twitter messages.
BIBREF152 provides a comparison of 8 off-the-shelf language identifiers applied without re-training to Twitter messages. One issue they report is that comparing the accuracy of off-the-shelf systems is difficult because of the different subset of languages supported by each system, which may also not fully cover the languages present in the target data. The authors choose to compare accuracy over the full set of languages, arguing that this best reflects the likely use-case of applying an off-the-shelf system to new data. They find that the best individual systems are , and , but that slightly higher accuracy can be attained by a simple voting-based ensemble classifier involving these three systems.
In addition to this, commercial or other closed-source language identifiers and language identifier services exist, of which we name a few. The Polyglot 3000 and Lextek Language Identifier are standalone language identifiers for Windows. Open Xerox Language Identifier is a web service with available REST and SOAP APIs.
Output:
| [
"what are the off-the-shelf systems discussed in the paper?"
] | task461-67798e7e5d2549559470a2da300a4829 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The average classification accuracy results are summarised in Table TABREF9.
Output:
| [
"What evaluation metric is used?"
] | task461-8650aab42b5b48d88797157e5919b184 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In this part, we discuss three significant limitations of BLEU and ROUGE. These metrics can assign: High scores to semantically opposite translations/summaries, Low scores to semantically related translations/summaries and High scores to unintelligible translations/summaries.
Challenges with BLEU and ROUGE ::: High score, opposite meanings
Suppose that we have a reference summary s1. By adding a few negation terms to s1, one can create a summary s2 which is semantically opposite to s1 but yet has a high BLEU/ROUGE score.
Challenges with BLEU and ROUGE ::: Low score, similar meanings
In addition not to be sensitive to negation, BLEU and ROUGE score can give low scores to sentences with equivalent meaning. If s2 is a paraphrase of s1, the meaning will be the same ;however, the overlap between words in s1 and s2 will not necessarily be significant.
Challenges with BLEU and ROUGE ::: High score, unintelligible sentences
A third weakness of BLEU and ROUGE is that in their simplest implementations, they are insensitive to word permutation and can give very high scores to unintelligible sentences. Let s1 be "On a morning, I saw a man running in the street." and s2 be “On morning a, I saw the running a man street”. s2 is not an intelligible sentence. The unigram version of ROUGE and BLEU will give these 2 sentences a score of 1.
Output:
| [
"What are the three limitations?"
] | task461-de07e1ed28b84a309c10cfd9c4af20a7 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We use 300 Dimensional Glove Common Crawl Embeddings (840B Token) BIBREF11 and fine tune them for the task
Output:
| [
"What embedding algorithm and dimension size are used?"
] | task461-88bffe587e144c66933393058e926d35 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Our definition of bias is now: [Interpretive Bias] An interpretive bias in an epistemic ME game is the probability distribution over types given by the belief function of the conversationalists or players, or the Jury. Note that in an ME game there are typically several interpretive biases at work: each player has her own bias, as does the Jury.
Output:
| [
"Which interpretative biases are analyzed in this paper?"
] | task461-c5de128213bc41559c0e3d2df0dfb3c8 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In terms of the future development of this technology, we emphasize the importance of answering four questions: 1) How to introduce unsupervised pre-training into NLG tasks with cross-modal context? 2) How to design a generic pre-training algorithm to fit a wide range of NLG tasks? 3) How to reduce the computing resources required for large-scale pre-training? 4) What aspect of knowledge do the pre-trained models provide for better language generation?
Output:
| [
"Which future direction in NLG are discussed?"
] | task461-2f96765ac55541ff81934725d502886e |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: First, we compare its performance with other BERT-models and state-of-the-art systems in sentiment analysis, to show its performance for classification tasks. Second, we compare its performance in a recent Dutch language task, namely the disambiguation of demonstrative pronouns, which allows us to additionally compare the zero-shot performance of our and other BERT models, i.e. using only the pre-trained model without any fine-tuning.
Output:
| [
"What language tasks did they experiment on?"
] | task461-4325cc2801fe4e3db7ff22c72bc0e18d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The overall performance of BERT was significantly better than the other models, having the lowest average difference in accuracy of 22.5 points.
Output:
| [
"Which model generalized the best?"
] | task461-9024b660e4d74ff5a7a51452c458823e |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We evaluate our proposed models on three commonly used knowledge graph datasets—WN18RR BIBREF26, FB15k-237 BIBREF18, and YAGO3-10 BIBREF27.
Output:
| [
"What benchmark datasets are used for the link prediction task?"
] | task461-dd7efb683d834ca286a62de0fe179544 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Embedding: We developed different variations of our models with a simple lookup table embeddings learned from scratch and using high-performance contextual embeddings, which are ELMo BIBREF11, BERT BIBREF16 and ClinicalBERT BIBREF13 (trained and provided by the authors).
Output:
| [
"What embeddings are used?"
] | task461-3f94320ec40c4b99a80aee4a2baa600b |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: These workers completed 1001 tasks: 496 tasks in the control and 505 in the AUI.
Output:
| [
"How many responses did they obtain?"
] | task461-2bcb546439744165b8114deb04674a38 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: A selection of results from this study are that:
Non-English code is a large-scale phenomena.
Transliteration is common in identifiers for all languages.
Languages clusters into three distinct groups based on how speakers use identifiers/comments/transliteration.
Non-latin script users write comments in their L1 script but write identifiers in English.
Right-to-left (RTL) language scripts, such as Arabic, have no observed prevalence on GitHub identifiers, implying that existing coders who speak RTL languages have substantial barriers in using their native script in code.
Output:
| [
"What are results of public code repository study?"
] | task461-76514a8851e44be1ad727b8156bf1f18 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Algorithm "Online Learning" demonstrates how our CNN model can be trained in a purely online setting. We first initialize the model parameters $\theta _0$ (line 1), which can be a trained model from other disaster events or it can be initialized randomly to start from scratch.
As a new batch of labeled tweets $B_t= \lbrace \mathbf {s}_1 \ldots \mathbf {s}_n \rbrace $ arrives, we first compute the log-loss (cross entropy) in Equation 11 for $B_t$ with respect to the current parameters $\theta _t$ (line 2a). Then, we use backpropagation to compute the gradients $f^{\prime }(\theta _{t})$ of the loss with respect to the current parameters (line 2b). Finally, we update the parameters with the learning rate $\eta _t$ and the mean of the gradients (line 2c). We take the mean of the gradients to deal with minibatches of different sizes. Notice that we take only the current minibatch into account to get an updated model.
Output:
| [
"What exactly is new about this stochastic gradient descent algorithm?"
] | task461-f86e060269f644c6807384ffe91bada2 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: An important feature when suggesting an article INLINEFORM0 to an entity INLINEFORM1 is the novelty of INLINEFORM2 w.r.t the already existing entity profile INLINEFORM3 Given an entity INLINEFORM0 and the already added news references INLINEFORM1 up to year INLINEFORM2 , the novelty of INLINEFORM3 at year INLINEFORM4 is measured by the KL divergence between the language model of INLINEFORM5 and articles in INLINEFORM6 . We combine this measure with the entity overlap of INLINEFORM7 and INLINEFORM8 . The novelty value of INLINEFORM9 is given by the minimal divergence value. Low scores indicate low novelty for the entity profile INLINEFORM10 .
Output:
| [
"What features are used to represent the novelty of news articles to entity pages?"
] | task461-13e9822539b94b73951007cf31deb9cc |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The straight line in figures FIGREF39, FIGREF43 and FIGREF51 is the result of a supervised LDA algorithm which is used as a baseline.
Output:
| [
"Was performance of the weakly-supervised model compared to the performance of a supervised model?"
] | task461-6459e7c98d2e4847800e2286a62e8777 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Our model improves the F1 score by almost 2%, which corresponds to a 12.3% error rate reduction.
Output:
| [
"How better is proposed model compared to baselines?"
] | task461-5775734a64174e99b5ad3f2d587ed72d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The above pseudocode is agnostic with respect to the choice of fragmentation and environment functions; task-specific choices are described in more detail for each experiment below.
Discussion
Output:
| [
"Which languages do they test on?"
] | task461-e4be43f930cf4c4fb949a699b1edd28b |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Among the GLUE tasks, these state-of-the-art systems make their greatest gains on the acceptability task with the Corpus of Linguistic Acceptability BIBREF0 . CoLA contains example sentences from linguistics publications labeled by experts for grammatical acceptability, and written to show subtle grammatical features. Because minimal syntactic differences can separate acceptable sentences from unacceptable ones (What did Bo write a book about? / *What was a book about written by Bo?), and acceptability classifiers are more reliable when trained on GPT and BERT than on recurrent models, it stands to reason that GPT and BERT have better implicit knowledge of syntactic features relevant to acceptability.
Output:
| [
"Where does the data in CoLA come from?"
] | task461-5c03d327094a465180c80e0db40d7164 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We consider two types of architectures: a bi-directional language model to augment the sequence to sequence encoder and a uni-directional model to augment the decoder. Both use self-attention BIBREF16 and the uni-directional model contains INLINEFORM0 transformer blocks, followed by a word classifier to predict the next word on the right. bi-directional language model to augment the sequence to sequence encoder
Output:
| [
"What language model architectures are used?"
] | task461-629a231f74e74a54a3b0d5a2b4f46095 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We introduce our proposed diversity, density, and homogeneity metrics with their detailed formulations and key intuitions.
Output:
| [
"Did they propose other metrics?"
] | task461-2a13a3d182a8438ab3b61e507dd566c3 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: SVMRank is a modification to SVM that assigns scores to each data point and allows the results to be ranked ( BIBREF26 ). We use SVMRank in the experiments below.
Output:
| [
"what is the supervised model they developed?"
] | task461-6d1b4d7a917b431bbcb47254a48bc4a0 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We used fastText and SVM BIBREF16 for preliminary experiments.
Output:
| [
"What classification models were used?"
] | task461-5ef1c5a7408a4c26bd510767b11a7d26 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: oss analysis. To conduct a more thorough evaluation, we further investigate model behaviours in terms of both reconstruc
Output:
| [
"How do they evaluate generated text quality?"
] | task461-b2a8ae2ed466442592254be754a88993 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Practical evaluation of GTD is currently only possible on synthetic data. We construct a range of datasets designed for image captioning evaluation. We call this diagnostic evaluation benchmark ShapeWorldICE (ShapeWorld for Image Captioning Evaluation). We illustrate the evaluation of specific image captioning models on ShapeWorldICE.
Output:
| [
"Are the images from a specific domain?"
] | task461-6ab6f159fd0c48b88a8b4a054712147d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Argus Dataset AI2-8grade/CK12 Dataset We consider this dataset as preliminary since it was not reviewed by a human and many hypotheses are apparently unprovable by the evidence we have gathered (i.e. the theoretical top accuracy is much lower than 1.0). MCTest Dataset We use an official extension of the dataset for RTE evaluation that again textually merges questions and answers.
Output:
| [
"what datasets did they use?"
] | task461-860cee3f0d6f4acaab1353d07f2e2576 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Just as introduced in sec:introduction, it is intractable to compute similarity exactly, as involving INLINEFORM0 computation. Hence, we consider the monte-carlo approximation: DISPLAYFORM0
where INLINEFORM0 is a list of entity pairs sampled from INLINEFORM1 . We use sequential sampling to gain INLINEFORM6 , which means we first sample INLINEFORM7 given INLINEFORM8 from INLINEFORM9 , and then sample INLINEFORM10 given INLINEFORM11 and INLINEFORM12 from INLINEFORM13 .
Output:
| [
"Which sampling method do they use to approximate similarity between the conditional probability distributions over entity pairs?"
] | task461-4ed1082cbc1043f589c25c8a764cd831 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: As with INLINEFORM0 , Greens-EFA, S&D, and EPP exhibit the highest cohesion, even though their ranking is permuted when compared to the ranking obtained with INLINEFORM1 . At the other end of the scale, we observe the same situation as with INLINEFORM2 . The non-aligned members NI have the lowest cohesion, followed by EFDD and ENL. The only place where the two methods disagree is the level of cohesion of GUE-NGL. The Alpha attributes GUE-NGL a rather high level of cohesion, on a par with ALDE, whereas the ERGM attributes them a much lower cohesion.
Output:
| [
"What insights does the analysis give about the cohesion of political groups in the European parliament?"
] | task461-48a8d49b23934b0897b0e57d26924d73 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: WikiSmall consists of ordinary and simplified sentences from the ordinary and simple English Wikipedias, which has been used as benchmark for evaluating text simplification BIBREF17 , BIBREF18 , BIBREF8 . The training set has 89,042 sentence pairs, and the test set has 100 pairs. WikiLarge is also from Wikipedia corpus whose training set contains 296,402 sentence pairs BIBREF19 , BIBREF20 . WikiLarge includes 8 (reference) simplifications for 2,359 sentences split into 2,000 for development and 359 for testing.
Output:
| [
"what are the sizes of both datasets?"
] | task461-cac755bbc949465e85ebb98d9cfc7ddc |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The UCL team had access to micro-narratives, as well as context specific meta-data such as demographic information and project details. For a cross-national comparison for policy-makers, the team translated the responses in multiple languages into English using machine translation, in this case Translate API (Yandex Technologies). As a pre-processing step, words without functional meaning (e.g. `I'), rare words that occurred in only one narrative, numbers, and punctuation were all removed. The remaining words were stemmed to remove plural forms of nouns or conjugations of verbs.
Output:
| [
"What elements of natural language processing are proposed to analyze qualitative data?"
] | task461-4706f583fa764fa0adc6f0480103d6aa |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In future work we also intend to add these types of studies to the ERP predictions.
Discussion
Output:
| [
"What datasets are used?",
"Which two pairs of ERPs from the literature benefit from joint training?"
] | task461-fea8f2db12c04918b7726843a70ed575 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We build and test our MMT models on the Multi30K dataset BIBREF21 . Each image in Multi30K contains one English (EN) description taken from Flickr30K BIBREF22 and human translations into German (DE), French (FR) and Czech BIBREF23 , BIBREF24 , BIBREF25 . The dataset contains 29,000 instances for training, 1,014 for development, and 1,000 for test. We only experiment with German and French, which are languages for which we have in-house expertise for the type of analysis we present. In addition to the official Multi30K test set (test 2016), we also use the test set from the latest WMT evaluation competition, test 2018 BIBREF25 .
Output:
| [
"Do they report results only on English dataset?"
] | task461-d8c0574ab2bd497f802e4fa7ffc04610 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: However, adversarial misspellings constitute a longstanding real-world problem. Spammers continually bombard email servers, subtly misspelling words in efforts to evade spam detection while preserving the emails' intended meaning BIBREF1 , BIBREF2 .
Output:
| [
"Why is the adversarial setting appropriate for misspelling recognition?"
] | task461-c8b518e8c1254f449c0e5f2911af8a8a |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Table 1 shows the results of all models on WikiLarge dataset. We can see that our method (NMT+synthetic) can obtain higher BLEU, lower FKGL and high SARI compared with other models, except Dress on FKGL and SBMT-SARI on SARI. It verified that including synthetic data during training is very effective, and yields an improvement over our baseline NMF by 2.11 BLEU, 1.7 FKGL and 1.07 SARI. Results on WikiSmall dataset are shown in Table 2. We see substantial improvements (6.37 BLEU) than NMT from adding simplified training data with synthetic ordinary sentences.
Output:
| [
"by how much did their model improve?"
] | task461-3d3c2045ea824875a5474856b71d5c81 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: To develop the dataset, we extracted 200 sentences from Sorani Kurdish books of grades one to three of the primary school in the Kurdistan Region of Iraq. We randomly created 2000 sentences from the extracted sentences.
Output:
| [
"What is the size of the dataset?",
"How was the dataset collected?"
] | task461-10a2db41f6e444caa3a9c1836bcc1017 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: First, it provides an annotation interface that allows users to define content elements, upload documents, and annotate documents Our platform is able to ingest documents in a variety of formats, including PDFs and Microsoft Word, and converts these formats into plain text before presenting them to the annotators.
Output:
| [
"What type of documents are supported by the annotation platform?"
] | task461-edce33c8f25d4ddb8461f50ca2195dc5 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Our Transformer Transducer model architecture has 18 audio and 2 label encoder layers. Every layer is identical for both audio and label encoders. The details of computations in a layer are shown in Figure FIGREF10 and Table TABREF11. All the models for experiments presented in this paper are trained on 8x8 TPU with a per-core batch size of 16 (effective batch size of 2048).
Output:
| [
"Does model uses pretrained Transformer encoders?"
] | task461-09a6a6becc854161ae9c00cdbc0db699 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Even among researchers familiar with the definitions outlined above, there was still a low level of agreement (Krippendorff's INLINEFORM0 ).
Output:
| [
"How was reliability measured?"
] | task461-a909a4bd2ce54e4684a5fe6a6a142ff4 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Our training data consists of 2.09M sentence pairs extracted from LDC corpus. To test different approaches on Chinese-to-English translation task, we use NIST 2003(MT03) dataset as the validation set, and NIST 2004(MT04), NIST 2005(MT05), NIST 2006(MT06) datasets as our test sets. For English-to-Chinese translation task, we also use NIST 2003(MT03) dataset as the validation set, and NIST 2008(MT08) will be used as test set.
Output:
| [
"What dataset did they use?"
] | task461-a940d4298ad8486883d95e93611b7779 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: In this paper, we provide a brief survey of existing datasets and describe the CRWIZ framework for pairing crowdworkers and having half of them acting as Wizards by limiting their dialogue options only to relevant and plausible ones, at any one point in the interaction Dialogue structure: we introduced structured dialogues through a Finite State Machine (FSM) that controls the current dialogue state and offers multiple suitable and relevant state transitions (actions) to the Wizard depending on the point in the interaction, the state of the world and the history. A graph of dialogue states, transitions and utterances is loaded when the system is initialised, and each chat room has its own dialogue state, which changes through actions.
Output:
| [
"How is dialogue guided to avoid interactions that breach procedures and processes only known to experts?"
] | task461-d58a5cf85f20413ca49c32cc5b7b8a1d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Particularly, it is not clear whether BERT shed light on solving tasks such as the Pronoun Disambiguation Problem (PDP) and Winograd Schema Challenge (WSC). On a Pronoun Disambiguation dataset, PDP-60, our method achieves 68.3% accuracy, which is better than the state-of-art accuracy of 66.7%. On a WSC dataset, WSC-273, our method achieves 60.3%.
Output:
| [
"Which datasets do they evaluate on?"
] | task461-aaf62d59a175432797f5367224e773dd |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: However, a significant bottleneck that has been overlooked by previous efforts is the input vocabulary size and its corresponding word embedding matrix, often accounting for a significant proportion of all model parameters. For instance, the embedding table of the BERTBASE model, comprising over 30K WordPiece tokens BIBREF14, accounts for over $21\%$ of the model size. While there has been existing work on reducing NLP model vocabulary sizes BIBREF15, distillation techniques cannot utilize these, since they require the student and teacher models to share the same vocabulary and output space. This profoundly limits their potential to further reduce model sizes.
Output:
| [
"Why are prior knowledge distillation techniques models are ineffective in producing student models with vocabularies different from the original teacher models? "
] | task461-e05a3485ff2b434f8754dc7e9355b3ce |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: CNN can also be employed on the sarcasm datasets in order to identify sarcastic and non-sarcastic tweets. We term the features extracted from this network baseline features, the method as baseline method and the CNN architecture used in this baseline method as baseline CNN. Since the fully-connected layer has 100 neurons, we have 100 baseline features in our experiment.
Output:
| [
"What are the network's baseline features?"
] | task461-5a5f739dbac44d629a7ca6b63a83dca9 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Inspired by the StackLite tag recommendation task on Kaggle, we build a new benchmark based on the public StackExchange data. We use questions with titles as source, and user-assigned tags as target keyphrases.
Since oftentimes the questions on StackExchange contain less information than in scientific publications, there are fewer keyphrases per data point in StackEx. Furthermore, StackExchange uses a tag recommendation system that suggests topic-relevant tags to users while submitting questions; therefore, we are more likely to see general terminology such as Linux and Java. This characteristic challenges models with respect to their ability to distill major topics of a question rather than selecting specific snippets from the text.
Output:
| [
"How was the StackExchange dataset collected?"
] | task461-4e056d71f1304ed5979f63c9d670c751 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: If we re-train the skip-gram model and update the emoji map periodically on new escort ads, when traffickers switch to new emojis, the map can link the new emojis to the old ones, assisting anti-trafficking experts in expanding the lexicon of trafficking flags. This approach also works for acronyms and deliberate typos.
Output:
| [
"How is the lexicon of trafficking flags expanded?"
] | task461-9f3867f814a24f38af34e00d2f8ff32f |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We also use the final hidden layer of the neural network as a task-specific embedding of the claim, together with the Web evidence.
Output:
| [
"What data is used to build the task-specific embeddings?"
] | task461-6e30f2b7397f456abb150ea771539087 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: For example, it can be used to extract features for usage with other machine learning tools, or to evaluate given features with the existing classifiers or regressors. Figure FIGREF19 demonstrates a simple feature extractor that retrieves the sentence length.
Output:
| [
"Do they show an example of usage for INFODENS?"
] | task461-9d16ed8a15ef4eca8c5594d05555b3ce |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Before designing this NLP toolkit, we conducted a survey among engineers and identified a spectrum of three typical personas.
Output:
| [
"How do the authors evidence the claim that many engineers find it a big overhead to choose from multiple frameworks, models and optimization techniques?"
] | task461-f037211239b74b8b9622f952d013a6c9 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: It is interesting to observe that the baseline model amplifies the bias in the training data set as measured by INLINEFORM0 and INLINEFORM1 . From measurements using the described bias metrics, our method effectively mitigates bias in language modelling without a significant increase in perplexity.
Output:
| [
"how is mitigation of gender bias evaluated?"
] | task461-eb715f2db0004651bb34d89d06af94d0 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Excluding the embedding weights, our model requires 100k parameters , while BIBREF8 requires 250k parameters.
Output:
| [
"how much was the parameter difference between their model and previous methods?",
"how many parameters did their model use?"
] | task461-484daa859dfa4cae816ac329e1c92444 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: For example, the original sentence `We went shop on Saturday' and the corrected version `We went shopping on Saturday' would produce the following pattern:
(VVD shop_VV0 II, VVD shopping_VVG II)
After collecting statistics from the background corpus, errors can be inserted into error-free text. The learned patterns are now reversed, looking for the correct side of the tuple in the input sentence. We only use patterns with frequency INLINEFORM0 , which yields a total of 35,625 patterns from our training data.
Output:
| [
"What textual patterns are extracted?"
] | task461-de4287793bf149e9af366a2443fa5c2d |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: WAS: The architecture used in BIBREF3 without the audio input. The decoder output Chinese character at each timestep. Others keep unchanged to the original implementation. LipCH-Net-seq: For a fair comparison, we use sequence-to-sequence with attention framework to replace the Connectionist temporal classification (CTC) loss BIBREF14 used in LipCH-Net BIBREF5 when converting picture to pinyin. CSSMCM-w/o video: To evaluate the necessity of video information when predicting tone, the video stream is removed when predicting tone and Chinese characters. In other word, video is only used when predicting the pinyin sequence. The tone is predicted from the pinyin sequence. Tone information and pinyin information work together to predict Chinese character.
Output:
| [
"What was the previous state of the art model for this task?"
] | task461-356c7f0800aa46f8b06d8e64fb172860 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: word embedding, input encoder, alignment, aggregation, and prediction
Output:
| [
"what is the architecture of the baseline model?"
] | task461-b756df30cb8846fcb159810f3067ad31 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: A single speaker narrated the 2000 sentences, which took several days.
Output:
| [
"How many annotators participated?"
] | task461-6d00c47a9f8d447d813dcc05cfc27e36 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Although techniques for constructing this knowledge base are outside the scope of this paper, we briefly mention them. Tables were constructed using a mixture of manual and semi-automatic techniques
Output:
| [
"How is the semi-structured knowledge base created?"
] | task461-db44e9a2f5c24228b551865a0e56ae1f |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We compare our approaches with related approaches of pivoting, multilingual NMT (MNMT) BIBREF19, and cross-lingual transfer without pretraining BIBREF16. The results show that our approaches consistently outperform other approaches across languages and datasets, especially surpass pivoting, which is a strong baseline in the zero-shot scenario that multilingual NMT systems often fail to beat BIBREF19, BIBREF20, BIBREF23.
Output:
| [
"which multilingual approaches do they compare with?"
] | task461-081f4b26677b4599a9ee5d61c266f70e |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Comparing these topics with those that appeared on a Catholic women forum, it seems that both ISIS and non-violent groups use topics about motherhood, spousal relationship, and marriage/divorce when they address women. Moreover, we used Depechemood methods to analyze the emotions that these materials are likely to elicit in readers. The result of our emotion analysis suggests that both corpuses used words that aim to inspire readers while avoiding fear. However, the actual words that lead to these effects are very different in the two contexts. Overall, our findings indicate that, using proper methods, automated analysis of large bodies of textual data can provide novel insight insight into extremist propaganda that can assist the counterterrorism community.
Output:
| [
"What conclusions do the authors draw from their finding that the emotional appeal of ISIS and Catholic materials are similar?"
] | task461-1f1de2296d7545f58409a261f67554f8 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: To show the effectiveness of our approach, we show results on the SICK dataset BIBREF1, a common benchmark for logic-based NLI, and find MonaLog to be competitive with more complicated logic-based approaches (many of which require full semantic parsing and more complex logical machinery).
Output:
| [
"Do they beat current state-of-the-art on SICK?"
] | task461-ff065fcef76d4b459bd1d5f98a8ff75e |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Although we create new models for tasks such as sentiment analysis and gender detection as part of AraNet, our focus is more on putting together the toolkit itself and providing strong baselines that can be compared to. Hence, although we provide some baseline models for some of the tasks, we do not explicitly compare to previous research since most existing works either exploit smaller data (and so it will not be a fair comparison), use methods pre-dating BERT (and so will likely be outperformed by our models) . For many of the tasks we model, there have not been standard benchmarks for comparisons across models. This makes it difficult to measure progress and identify areas worthy of allocating efforts and budgets.
Output:
| [
"What models did they compare to?"
] | task461-9faf7fc1c19c46fd8df1b7412463e438 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: The participants are asked to score each summary on three indicators: relevance, conciseness and readability on a 1 to 5 scale, and rank the summary pair (tie allowed). We observe that SPNet reaches the highest score in both ROUGE and CIC
Output:
| [
"What automatic and human evaluation metrics are used to compare SPNet to its counterparts?"
] | task461-7e193850fe5847b887dd7cc6d956c5b6 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: We compare our proposed discrete CVAE (DCVAE) with the two-stage sampling approach to three categories of response generation models:
Baselines: Seq2seq, the basic encoder-decoder model with soft attention mechanism BIBREF30 used in decoding and beam search used in testing; MMI-bidi BIBREF5, which uses the MMI to re-rank results from beam search.
CVAE BIBREF14: We adjust the original work which is for multi-round conversation for our single-round setting. For a fair comparison, we utilize the same keywords used in our network pre-training as the knowledge-guided features in this model.
Other enhanced encoder-decoder models: Hierarchical Gated Fusion Unit (HGFU) BIBREF12, which incorporates a cue word extracted using pointwise mutual information (PMI) into the decoder to generate meaningful responses; Mechanism-Aware Neural Machine (MANM) BIBREF13, which introduces latent embeddings to allow for multiple diverse response generation.
Output:
| [
"What other kinds of generation models are used in experiments?"
] | task461-b36a7682869e4e87bbe003f9e9a21da0 |
Definition: In this task, you will be presented with a context from an academic paper and you have to write an answerable question based on the context. Your questions can be extractive, abstractive, or yes-no questions.
Positive Example 1 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: how was the dataset built?
Positive Example 2 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language are the tweets?
Negative Example 1 -
Input: Using this annotation model, we create a new large publicly available dataset of English tweets.
Output: In what language is tweets?
Negative Example 2 -
Input: Questions are gathered from anonymized, aggregated queries to the Google search engine. Queries that are likely to be yes/no questions are heuristically identified: we found selecting queries where the first word is in a manually constructed set of indicator words and are of sufficient length, to be effective.
Questions are only kept if a Wikipedia page is returned as one of the first five results, in which case the question and Wikipedia page are given to a human annotator for further processing.
Annotators label question/article pairs in a three-step process. First, they decide if the question is good, meaning it is comprehensible, unambiguous, and requesting factual information. This judgment is made before the annotator sees the Wikipedia page. Next, for good questions, annotators find a passage within the document that contains enough information to answer the question. Annotators can mark questions as “not answerable" if the Wikipedia article does not contain the requested information. Finally, annotators mark whether the question's answer is “yes" or “no". Annotating data in this manner is quite expensive since annotators need to search entire Wikipedia documents for relevant evidence and read the text carefully.
Output: What is the size of the dataset?
Now complete the following example -
Input: Dataset: We perform experiments on two widely-used tasks for the English-to-Japanese language pair: KFTT BIBREF12 and BTEC BIBREF13 . KFTT is a collection of Wikipedia article about city of Kyoto and BTEC is a travel conversation corpus. BTEC is an easier translation task than KFTT, because KFTT covers a broader domain, has a larger vocabulary of rare words, and has relatively long sentences.
Output:
| [
"What datasets were used?"
] | task461-9d00f5d4d4034bc098b8e928e7e412a3 |