Interactive Medical Diagnosis on RareArena
59.9ICRQwen2.5-72B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-72BInteractive Strategy=REFINE2026.01 | 59.9 | 17 | |
| GPT-5-miniInteractive Strategy=REFINE2026.01 | 51.8 | 32 | |
| Qwen2.5-72BInteractive Strategy=SC2026.01 | 50.1 | 15.5 | |
| Qwen2.5-72BInteractive Strategy=ReAct2026.01 | 47.4 | 13 | |
| GPT-5-miniInteractive Strategy=SC2026.01 | 46.2 | 30.5 | |
| Qwen2.5-72BInteractive Strategy=Baseline2026.01 | 41.6 | 10.5 | |
| Qwen2.5-7BInteractive Strategy=Baseline2026.01 | 41.2 | 5 | |
| GPT-5-miniInteractive Strategy=ReAct2026.01 | 39.3 | 24 | |
| Qwen2.5-7BInteractive Strategy=REFINE2026.01 | 38.2 | 7.5 | |
| Qwen2.5-7BInteractive Strategy=SC2026.01 | 37 | 5.5 | |
| GPT-5-miniInteractive Strategy=Baseline2026.01 | 35.7 | 15.5 | |
| Qwen2.5-7BInteractive Strategy=ReAct2026.01 | 29.8 | 2.5 |