ResidencyRL: RL in simulated clinics nearly doubles psychiatric diagnostic appropriateness to 68%
SRSchmidgall · x · 2026-10-31
A new preprint from Google researchers introduces ResidencyRL, training clinical AI agents via multi-turn simulated encounters (up to 60 dialogue turns, 8 tool calls per trajectory). Applied to psychiatry, an agent trained purely against simulated patients was preferred over its base model in 76.7% of scenario pairs in a double-blind OSCE with 32 role-playing psychiatrists, nearly doubling diagnostic appropriateness from 33% to 68% by learning to ask better questions.
More from Research
- NVIDIA Nemotron 3.5 scales LLM self-verification, +9-10 points on agentic benchmarks — sarahcat21 · 2026-10-31
- Unofficial OpenTargets MCP Server Brings Gene-Drug-Disease Data to LLMs — modelcontextprotocol · 2026-10-31
- Researchers find 5 personality structures in LLM weight space, Big Five-like — luke_drago_ · 2026-10-31
- NVS benchmarks are misaligned: proper pose alignment puts 3D Gaussians back at SOTA — kwangmoo_yi · 2026-10-31
- arXiv moderator: submission surge causes months-long holds, new moderators wanted — tdietterich · 2026-10-31
- arXiv policy checks reject class projects, proposals, surveys and position papers — tdietterich · 2026-10-31