Scrape arXiv listings and rank new papers by your own keywords, authors and subject weights — Streamlit GUI plus CLI
See the code
arXiv Digest reads the new papers in the arXiv categories you follow and ranks them by your own keywords, authors and categories, so the papers you care about come first.
quant-ph, hep-th, cs.LG, math.AG, …You need uv
(curl -LsSf https://astral.sh/uv/install.sh | sh on macOS and Linux).
git clone https://github.com/isaac-tes/arxiv-digest.git
cd arxiv-digest
uv tool install '.[gui]'
arxiv-gui # opens the web app at http://localhost:8501
Click Fetch papers, then start from a
starter preset or type your own keywords and
authors. The list re-ranks as you edit. Update later with arxiv-digest update.
Watch it as full-quality video
arxiv-digest --top 10 prints the same ranking in the terminal,
or writes Markdown and JSON for cron jobs and notes apps. It reads the same
arxiv_config.json as the web app. See the CLI guide.| Quickstart | Install options, updating, running from a clone |
| GUI guide | Every tab, highlighting, Zotero, profiles and presets |
| Scoring | The rules, default weights and how to tune them |
| CLI guide | All flags, the config file, output formats |
| iPhone app | Install, modes, web ↔ iOS feature table |
| Sync across devices | Run the digest server, connect your devices |
| Troubleshooting | Common problems and fixes |
Or browse them all on the documentation site.
AutoLLM/ArxivDigest asks GPT how relevant each paper is to a description of your interests and emails you the result. It needs an OpenAI key and a SendGrid account and hasn't been updated since May 2024. arXiv Digest is unrelated to it and shares no code: it scores with rules you can read and change, runs locally on demand, and adds the browser and iPhone apps and Zotero saving. Both are MIT-licensed.
Bug reports and pull requests are welcome; see CONTRIBUTING.md.
uv sync --group dev && uv run pytest runs the test suite (no network needed).
Scrape arXiv listings and rank new papers by your own keywords, authors and subject weights — Streamlit GUI plus CLI
See the code
arXiv Digest reads the new papers in the arXiv categories you follow and ranks them by your own keywords, authors and categories, so the papers you care about come first.
quant-ph, hep-th, cs.LG, math.AG, …You need uv
(curl -LsSf https://astral.sh/uv/install.sh | sh on macOS and Linux).
git clone https://github.com/isaac-tes/arxiv-digest.git
cd arxiv-digest
uv tool install '.[gui]'
arxiv-gui # opens the web app at http://localhost:8501
Click Fetch papers, then start from a
starter preset or type your own keywords and
authors. The list re-ranks as you edit. Update later with arxiv-digest update.
Watch it as full-quality video
arxiv-digest --top 10 prints the same ranking in the terminal,
or writes Markdown and JSON for cron jobs and notes apps. It reads the same
arxiv_config.json as the web app. See the CLI guide.| Quickstart | Install options, updating, running from a clone |
| GUI guide | Every tab, highlighting, Zotero, profiles and presets |
| Scoring | The rules, default weights and how to tune them |
| CLI guide | All flags, the config file, output formats |
| iPhone app | Install, modes, web ↔ iOS feature table |
| Sync across devices | Run the digest server, connect your devices |
| Troubleshooting | Common problems and fixes |
Or browse them all on the documentation site.
AutoLLM/ArxivDigest asks GPT how relevant each paper is to a description of your interests and emails you the result. It needs an OpenAI key and a SendGrid account and hasn't been updated since May 2024. arXiv Digest is unrelated to it and shares no code: it scores with rules you can read and change, runs locally on demand, and adds the browser and iPhone apps and Zotero saving. Both are MIT-licensed.
Bug reports and pull requests are welcome; see CONTRIBUTING.md.
uv sync --group dev && uv run pytest runs the test suite (no network needed).