RAG-Pipeline for Evidence Extraction from Scientific Articles

This repository contains the data and source code for the paper submission titled "Leveraging Language Models to Discover Evidence-Based Actions for OSS Sustainability."

• The ReACTGPT directory includes the code for the RAG pipeline. Refer to the readme.md file for detailed instructions on how to run the project.

• The Notebooks directory contains various pre-processing and post-processing notebooks.

• The DATASET directory holds all relevant files related to the project.