A customized AI papers rss feed for personal use.
User can filter the papers related to the field of interest using api calls.
Summarization of each paper and explanation why the paper is selected are provided in the feed.
User can subscribe to the following feeds:
- arxiv-cs-AI
- hf
pip install -r requirements.txt
Also we need to install Redis on our local machine.
Port number is set to 6379 in redis_client.py.
Edit the config.py file to set the API key of the LLM.
Using SiliconFlow Qwen-7b instruct API as default base url because it is free.
python server.py
Subscribe with the following url: Arxiv:
https://127.0.0.1:5000/feed-arxiv # xml
https://127.0.0.1:5000/feed-arxiv-json # json
Hugging Face:
https://127.0.0.1:5000/feed-hf # xml
https://127.0.0.1:5000/feed-hf-json # json
User can customize base rss url, the field of interest, language for answering, prompt for filtering and the number of papers to fetch by modifying the config.py file.
- add more feeds support
- add more source llm support