This project provides a Telegram bot that delivers daily arXiv paper recommendations based on user-defined keywords and feedback. It targets researchers and academics seeking to stay updated with relevant literature, offering a personalized and interactive way to discover new papers.

How It Works

The bot fetches papers from arXiv using specified keywords. Users interact via Telegram, rating papers with emojis (:thumbsup:, :thumbsdown:, :heart:) which are logged. These ratings, along with user-submitted paper links and tags, are used to retrain a preference model, enhancing future recommendations.

Quick Start & Requirements

• Install dependencies: pip install -r requirement.txt
• Run the bot: python arxiv_checker.py --first_backcheck_day 3 --keywords llm,search,reasoning,planning,optimization
• Requires Telegram Bot Token, Chat ID, Telegram API ID, and API Pass, configured as environment variables.
• Official instructions for obtaining Telegram API credentials are linked in the README.

Highlighted Details

• Daily arXiv paper recommendations via Telegram.
• Interactive rating system using emojis (:thumbsup:, :thumbsdown:, :heart:).
• Supports user-suggested papers and tag-based retrieval.
• Model retraining mechanism based on user feedback.

Maintenance & Community

No specific community channels or notable contributors are mentioned in the README.

Licensing & Compatibility

MIT License. Permissive for commercial use and integration with closed-source projects.

Limitations & Caveats

The bot requires manual setup of Telegram API credentials and environment variables. Model retraining involves separate script execution. The README does not detail the specific machine learning models used or their performance benchmarks.