Blinded A/B preference on Clinician Obesity cohort R1 (val)
82.2Win PercentageCLR-voyance-8B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CLR-voyance-8Bbaseline_comparison=HuatuoGPT-o1-7B2026.05 | 82.2 | 76 | |
| CLR-voyance-8Bbaseline_comparison=Qwen3-8B (base)2026.05 | 62.7 | 56 | |
| CLR-voyance-8Bbaseline_comparison=GPT-52026.05 | 48 | 41 |