DiscoveryRSI: recursive multi-agent evolution for scientific hypothesis discovery
PhAILabs · hf · 2026-10-19
DiscoveryRSI, from PhAI Labs and 16 partner institutions including CUHK, Michigan, Stanford, and Harvard Medical School, uses multi-agent evolutionary search and recursive self-improvement to generate testable scientific hypotheses, outperforming baseline agents on life-science benchmarks. Code, models, and the DiscoveryRSI Arena leaderboard are publicly released.
More from Research
- MBZUAI Releases Jais 3: Six Open Arabic Models from 0.9B to 375B with Full Training Recipe — lmoroney · 2026-10-19
- Dietterich explains arXiv's one-shot policy for major results — tdietterich · 2026-10-19
- DiscoveryRSI paper: 27B model's hypotheses hit 6.3%, beating GPT-6 and Claude Opus 5 — Afinetheorem · 2026-10-19
- GeoState: View-Decoupled Geometry States Separate Scene Dynamics from Camera Motion in World Models — Tongji · 2026-10-19
- OS-Helix: End-to-End Agent Unifies GUI Actions and Code Execution for Desktop Automation — ustc · 2026-10-19
- Sparse Autoencoders with IDF Weighting Match Symbols Across Systems, Winning 124 of 129 Queries — AI-Interpretability-illinois · 2026-10-19