AutoScientists NeurIPS paper: self-organizing AI research teams hit 74.4 percentile on BioML-Bench
marinkazitnik · x · 2026-09-25
AutoScientists, a NeurIPS-accepted paper from Shanghua Gao, Ada Fang, and Marinka Zitnik, introduces a decentralized team of AI agents for long-running computational scientific experimentation.
- No central planner or fixed workflow: agents self-organize into teams around promising directions, explore in parallel, and reorganize as evidence emerges
- A discussion phase critiques and filters proposals before committing compute; teams share both successes and failures to avoid redundant exploration
Results under matched budgets: 74.4% mean leaderboard percentile across 24 BioML-Bench tasks (imaging, protein engineering, single-cell omics, drug discovery), +8.33% over the strongest prior biomedical agent; on LLM training optimization it reaches the target validation bits-per-byte 1.9× faster than autoresearch. Paper and code are public.
More from coding & agent
- SkillRL (NeurIPS 2026): 7B model beats GPT-4o by 41% via recursive skill evolution — cihangxie · 2026-09-25
- Weave Code Max offers $50+ of coding model usage for $10/month — ycombinator · 2026-09-25
- Mad science: Jev autopilot lands a plane in a terminal flight simulator — bilawalsidhu · 2026-09-25
- Greptile reviewed 395K PRs at NVIDIA, cutting merge time from 24h to 6h — ycombinator · 2026-09-25
- Open-source Claude skill turns one prompt into stop-motion claymation films in Blender — angrypenguinPNG · 2026-09-25
- Descript launches MCP to edit video from inside Claude, ChatGPT, and Codex — descript · 2026-09-25