Interactive Medical Diagnosis on DiagnosisArena
73.8ICRQwen2.5-72B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-72BInteractive Strategy=REFINE2026.01 | 73.8 | 18.5 | |
| GPT-5-miniInteractive Strategy=REFINE2026.01 | 64.6 | 42 | |
| Qwen2.5-72BInteractive Strategy=SC2026.01 | 63.4 | 15.5 | |
| GPT-5-miniInteractive Strategy=SC2026.01 | 60.8 | 39.5 | |
| Qwen2.5-72BInteractive Strategy=ReAct2026.01 | 58.9 | 12.5 | |
| Qwen2.5-72BInteractive Strategy=Baseline2026.01 | 55.5 | 10 | |
| GPT-5-miniInteractive Strategy=ReAct2026.01 | 53.9 | 29.5 | |
| Qwen2.5-7BInteractive Strategy=SC2026.01 | 53.6 | 10.5 | |
| Qwen2.5-7BInteractive Strategy=Baseline2026.01 | 53.1 | 8 | |
| Qwen2.5-7BInteractive Strategy=REFINE2026.01 | 50.8 | 13 | |
| GPT-5-miniInteractive Strategy=Baseline2026.01 | 47.2 | 23 | |
| Qwen2.5-7BInteractive Strategy=ReAct2026.01 | 45.1 | 3.5 |