chansung commited on
Commit
a06f38e
·
verified ·
1 Parent(s): 34e800c

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +24 -0
  2. configs/prompts.toml +11 -8
README.md CHANGED
@@ -59,3 +59,27 @@ $ python main.py # or gradio main.py
59
 
60
  # Acknowledgments
61
  This is a project built during the Vertex sprints held by Google's ML Developer Programs team. We are thankful to be granted good amount of GCP credits to do this project.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59
 
60
  # Acknowledgments
61
  This is a project built during the Vertex sprints held by Google's ML Developer Programs team. We are thankful to be granted good amount of GCP credits to do this project.
62
+ # AdaptSum
63
+
64
+ AdaptSum stands for Adaptive Summarization. This project focuses on developing an LLM-powered system for dynamic summarization. Instead of generating entirely new summaries with each update, the system intelligently identifies and modifies only the necessary parts of the existing summary. This approach aims to create a more efficient and fluid summarization process within a continuous chat interaction with an LLM.
65
+
66
+ # Instructions
67
+
68
+ 1. Install dependencies
69
+ ```shell
70
+ $ pip install requirements.txt
71
+ ```
72
+
73
+ 2. Setup Gemini API Key
74
+ ```shell
75
+ $ export GEMINI_API_KEY=xxxxx
76
+ ```
77
+ > note that GEMINI API KEY should be obtained from Google AI Studio. Vertex AI is not supported at the moment (this is because Gemini SDK does not provide file uploading functionality for Vertex AI usage now).
78
+
79
+ 3. Run Gradio app
80
+ ```shell
81
+ $ python main.py # or gradio main.py
82
+ ```
83
+
84
+ # Acknowledgments
85
+ This is a project built during the Vertex sprints held by Google's ML Developer Programs team. We are thankful to be granted good amount of GCP credits to do this project.
configs/prompts.toml CHANGED
@@ -1,20 +1,23 @@
1
  [summarization]
2
  prompt = """
3
  Below is the initial summary of our conversation.
4
- Based on the summary and the last conversation between you (assistant) and me (user), I want to update the summary.
5
 
6
- **Initial Summary:**
7
  $previous_summary
8
 
9
  **Last Conversation:**
10
  $latest_conversation
11
 
12
- When updating the summary:
13
- * **Focus:** Only include information we have explicitly discussed in this session. Do not introduce any new information or topics, even if you have prior knowledge.
14
- * **Accuracy:** Ensure the summary is factually accurate and reflects the nuances of our discussion.
15
- * **Completeness:** Strive to be as comprehensive and detailed as possible, capturing all key points and insights.
16
- * **Conciseness:** While being detailed, also aim for conciseness and clarity in the summary.
17
- * **Update Strategy:** Instead of rewriting the entire summary each time, update only the specific portions necessary to reflect new information or changes in understanding.
 
 
 
18
  """
19
 
20
  system_prompt = """
 
1
  [summarization]
2
  prompt = """
3
  Below is the initial summary of our conversation.
4
+ Based on the given summary and the last conversation between you (assistant) and me (user), I want to update the summary.
5
 
6
+ **Summary:**
7
  $previous_summary
8
 
9
  **Last Conversation:**
10
  $latest_conversation
11
 
12
+ Summary Guide:
13
+ * Do not rewrite the entire summary.
14
+ * It is allowed to modify the current summary, but do not eliminate.
15
+ * Update only the specific portions necessary to reflect new information or changes.
16
+ * Only include information we have explicitly discussed in this session. Do not introduce any new information or topics, even if you have prior knowledge.
17
+ * Ensure the summary is factually accurate and reflects the nuances of our discussion.
18
+ * While being detailed, also aim for conciseness and clarity in the summary.
19
+ * Use markdown formatting to make the summary more readable
20
+ * Do not seprate sections for previous and updated summaries
21
  """
22
 
23
  system_prompt = """