Interactive Medical Diagnosis on ClinicalBench
0.511ICRQwen2.5-72B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-72BInteractive Strategy=REFINE2026.01 | 0.511 | 0.53 | |
| GPT-5-miniInteractive Strategy=REFINE2026.01 | 0.435 | 0.495 | |
| Qwen2.5-72BInteractive Strategy=SC2026.01 | 0.432 | 0.495 | |
| GPT-5-miniInteractive Strategy=SC2026.01 | 0.415 | 0.49 | |
| Qwen2.5-7BInteractive Strategy=Baseline2026.01 | 0.41 | 0.355 | |
| Qwen2.5-72BInteractive Strategy=Baseline2026.01 | 0.408 | 0.435 | |
| Qwen2.5-72BInteractive Strategy=ReAct2026.01 | 0.387 | 0.42 | |
| Qwen2.5-7BInteractive Strategy=REFINE2026.01 | 0.353 | 0.385 | |
| GPT-5-miniInteractive Strategy=ReAct2026.01 | 0.335 | 0.495 | |
| Qwen2.5-7BInteractive Strategy=SC2026.01 | 0.325 | 0.35 | |
| GPT-5-miniInteractive Strategy=Baseline2026.01 | 0.313 | 0.465 | |
| Qwen2.5-7BInteractive Strategy=ReAct2026.01 | 0.298 | 0.33 |