Paper2Agent turns research papers into interactive AI agents
Paper2Agent is an open‑source framework that automatically converts academic papers, their code and data into runnable AI agents. Stanford computer scientist James Zou and his team described the system in a Nature paper on September 16 and demonstrated it on the AlphaGenome deep‑learning model, which predicts DNA‑mutation effects. Feeding the AlphaGenome documentation into Paper2Agent produced…
Key points
- Paper2Agent generated 22 AlphaGenome tools in about 45 minutes for under $15 compute cost.
- Of 100 computational‑biology papers tested, 26 could not be converted into agents due to incomplete code or documentation.
- Agents collaborated to identify gene GPR137 as a psoriasis candidate and suggested ten validation methods.
The researchers also linked agents built from two other papers with the AlphaGenome agent, allowing the trio to pinpoint the gene GPR137 as a likely causal factor for psoriasis and to propose ten validation experiments, one of which a human selected and confirmed. In a broader test of 100 computational‑biology papers, 26 could not be turned into agents because of missing code or documentation, a failure the authors view as a useful diagnostic. The team has already used Paper2Agent to create a “Virtual Biotech” multi‑agent platform and even generated an agent for the Paper2Agent paper itself, hosted at paper2agent.ai.
Why Read a Research Paper When You Can Turn It Into an AI Agent?
IEEE Spectrum AI · 22 September 2026
Loading the full article…
This text was published by IEEE Spectrum AI and written by Elie Dolgin. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- Amazon adds skill selection and instruction-following evaluators to Bedrock AgentCore · 1 src
- Meta tests Muse calls that are actually made by humans in a call center · 1 src
- Meta's Muse AI agent tops US App Store, drives stock surge · 16 src
- Claude Code stops working, user fixes it with Google's Antigravity (Gemini) · 1 src
- Claude Code adds support for OpenAI Agents.md format · 3 src
Comments
via GitHub Discussions