Simulation-trained Gemini nearly doubles clinical rubric score, 0.706 vs 0.379
SRSchmidgall · x · 2026-10-31
Across 127 simulated intake scenarios validated by 54 board-certified psychiatrists, the simulation-RL-trained agent nearly doubled its base model's clinical rubric score (0.706 vs. 0.379; composite 0.564 vs. 0.294), with gains preserved under an independent out-of-family judge (GPT-5.6-Luna).
The evaluation framework shows licensed psychiatrists grading scenario quality on eight criteria (clinical realism, diagnostic coherence, difficulty calibration, etc.), and a radar chart shows the trained model beating the base model across diagnostic appropriateness, therapeutic alliance, clinical management, and documentation.
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