Welcome to the RAG System for Contract Q&A! This repository contains the codebase and resources for building, evaluating, and improving a Retrieval Augmented Generation (RAG) system for Contract Question & Answering (Q&A). The goal is to develop a powerful contract assistant that can autonomously answer questions about contracts, with the ultimate aim of creating a fully autonomous contract bot capable of drafting, reviewing, and negotiating contracts independently.
Retrieval Augmented Generation (RAG) is a hybrid AI model that combines the power of large language models with external data sources. RAG leverages a large language model to generate responses but first retrieves relevant information from external data sources, enabling it to provide more accurate and context-rich outputs.
The repository contains an evaluation set with two contracts (a short one and a long one), each with a list of ten questions and their correct answers.
- /code: Contains the codebase for the RAG system.
- /data: Contains the evaluation set and any additional datasets used.
- Clone the repository:
git clone https://github.com/AbelBekele/Contract-Advisor-RAG-.git
- Install dependencies:
pip install -r requirements.txt
- Explore the codebase to understand the RAG system and evaluation set.
- Run the RAG system on the evaluation set and analyze the results.
- Experiment with different techniques to improve RAG performance.
Contributions are welcome! If you have ideas for improving the RAG system or resources to add, please submit a pull request.
This repository is licensed under the MIT License. See the LICENSE file for details.
- The creators of the Langchain and Chroma courses for their informative resources.
- OpenAI for their contributions to AI research and development.
For any inquiries or feedback, please contact [email protected].