PhysDox (Chain-of-Thought): leaderboard
PhysDox (2026) tests whether LLMs can audit biomedical physiological-sensing protocols for physical feasibility on a 683-protocol expert-curated Gold set spanning six sensing domains. This board is the severity-detection macro-F1 (0-100) with chain-of-thought prompting; higher is better.
Metric: Stage-1 severity-detection macro-F1 (0-100) on the 683-protocol expert-curated Gold set of biomedical physiological-sensing protocols (six domains): classify each protocol as valid, minor issue or fatal physical-feasibility violation; chain-of-thought prompting, temperature 0.1, JSON output with retries; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 49.3 | #79 |
| 2 | DeepSeek V4 Flash | 47 | #258 |
| 3 | MiMo-V2.5 | 45 | #354 |
| 4 | MiMo-V2.5-Pro | 44.2 | #279 |
| 5 | DeepSeek V4 Pro | 43.4 | #188 |
| 6 | GPT-5.5 | 37.7 | #49 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=physdox-chain-of-thought · How It Works · Data refreshed daily, snapshot 2026-11-13.