DA ‐ Knowledge Distiller - nerority/Advanced-GPTs GitHub Wiki

Automated Knowledge Distillation

Public GPT Link: Access GPT Here

Description

This GPT functions as an AI Knowledge Distiller, specifically designed to manage and distill knowledge from user-uploaded documents into comprehensive summaries. It oversees a sophisticated workflow involving data acquisition, strategic analysis, and synthesis of knowledge into structured summaries.

Usage Instructions

Users start by uploading a document for knowledge distillation. The tool processes this input through a series of steps, involving comprehensive searches, focused searches, and synthesis of key information. The entire process is managed within a dataframe to track changes and improvements iteratively.

User Commands

  • !start: Initiates the workflow based on the user-uploaded document.
  • !demo: Demonstrates the complete workflow using AI synthesized data.
  • !dive [specific areas of interest]: Executes a follow-up workflow to complete a deep dive into user-specified areas.
  • !help: Provides guidance on how to use commands.
  • G: Generates the synthesized summary outside of the Python tool in well-formatted markdown.

🌟 Important Notes

  • This tool is essential for researchers, analysts, and students who need to distill large documents into concise, structured knowledge summaries.
  • The workflow includes comprehensive searches, focused analysis, and synthesis steps to ensure the extracted knowledge is well-organized and actionable.

Workflow

sequenceDiagram
    participant User
    participant ChatGPT
    participant Python
    participant RAG

    User->>ChatGPT: Uploads Document (!start)
    ChatGPT->>Python: Knowledge Distillation Module
    Python-->>ChatGPT: Return
    ChatGPT->>Python: Plan Workflow Sequence
    Python-->>ChatGPT: Workflow Tool Sequence
    ChatGPT->>RAG: Conducts Initial Document Search
    RAG-->>Python: Transfers Document Understanding
    ChatGPT->>Python: Create Strategy for Knowledge Distillation
    Python-->>ChatGPT: Outlines Document Analysis Strategy
    loop RAG Search Loop for Key Concepts
        ChatGPT->>RAG: Performs Focused Searches
        RAG-->>Python: Delivers Search Results
        ChatGPT->>Python: Synthesize Key Information
        Python-->>ChatGPT: Updated Key Information
    end
    ChatGPT->>Python: Compile Comprehensive Knowledge Summary
    Python-->>ChatGPT: Synthesized Summary and Save
    ChatGPT->>User: Displays Save Link and Commands
    User->>ChatGPT: Issues "G" to Generate the Synthesized Summary
    ChatGPT->>User: Full Synthesized Summary

Example Usage