DeepMind's ResidencyRL: simulated RL lifts psychiatric intake AI, preferred in 76.7% of cases
SRSchmidgall · x · 2026-10-31
A Google DeepMind/Google Research preprint extends ResidencyRL to psychiatric intake. Multi-turn RL trained only against simulated patients made the agent preferred over its identically prompted base model in 76.7% of scenario pairs in a double-blind, psychiatrist-enacted OSCE (N=60 pairs, 120 encounters).
- Gemini-generated patient scenarios with quality verification feed a GRPO-based RL loop with clinical rubrics and autoraters.
- Adversarial safety failures dropped sharply: missed red flags fell from 10.9% to 6.2%, under-triage from 30.5% to 12.9%.
- Authors stress simulation is the training ground, but evidence must ultimately come from clinicians and patients.
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