ChatArxiv is a project that uses ChatGPT as an Arxiv paper reading assistant. With ChatArxiv, users can quickly understand and digest academic papers on Arxiv, reducing the time spent on reading papers. Users simply need to provide their OpenAI API key and the corresponding Arxiv article link to interact with ChatGPT (ArxivBot) and gain in-depth knowledge about the paper's details.
Install Dependency
pip install -r requirements.txt
Launch Server
python app.py
Basic Usage
Input OpenAI API key and arxiv link,push set basic button and here you go. You can also select multiple sections for a more detailed conversation.
Currently in the very early stages of development, here are the features we plan to add in future versions:
- Language switching: Allowing users to switch the interface and chat language according to their needs.
- Style switching: Providing users with multiple chat style options to meet different reading and learning needs.
- Image reading: Enabling ChatGPT to recognize and interpret images, charts, and other visual content in papers.
- PDF preview: Previewing the full PDF version of the paper directly within ChatArxiv.
- Paper history management: Allowing users to manage the papers they have read for easy review and further research.