arXiv Research Paper (PDF) Summarization Tool

This repository houses a Python tool designed to process and summarize arXiv research papers (PDF files), making them more accessible and easier to digest. This tool creates a summarized report (a PDF file) of the research making the original research paper easier to read and more accessible. This tool is intended for education and informational purposes only and original IP remains with the author of the research paper.

Features

• PDF Reader: Reads PDF files, with the option to exclude specific pages as provided by the user (references and some other pages towards the end of the report).
• OpenAI Summarizer: A Summarizes the research paper by creating smaller chunks and then summarizing each chunk and then consolidating all the part summaries to create a final overall summary. The tool uses OpenAI's GPT 4 model and requires a paid OpenAI API key.
• PDF Writer: Creates or appends Summarized report to a PDF files specified by the user.

How to use - Input and output file set up

• Setting up input file: default input pdf file path/directory is paper_pdf_repo/research_papers. Save the file in that dir.
• Setting up output file: default output folder where pdf file with summarized report is created is paper_pdf_repo/summaries
• Setting up OpenAI API: Rename the .env-example file to .env file and update it with your OpenAI API key.
• Install required libraries: use pip install -r requirements.txt to install required python packages to run the program.
• Before Using: update the input file name, output file name, and pages to exclude (generally last few pages with references etc.) in summarize_arxiv_paper.py (Will add this to streamlit UI in future).
• How to run: After setting up API key, updating the pages to be excludes, and adjusting file name, execute summarize_arxiv_paper.py to run the tool.

Upcoming Features

• Open Source Model Summarizer: Create the workflow using open source model to summarize research papers free of cost.
• User Interface: Create a streamlit user interface.
• process pdf from arxiv URL: No need to download the pdf file, user can just provide the url to the pdf file.