Medical on MedXpertQA
15.25AccuracyINFUSER
Evaluation Results
| Method | Links | |
|---|---|---|
| INFUSERModel Backbone=Qwen3-8B-Base2026.06 | 15.25 | |
| R-ZeroModel Backbone=Qwen3-8B-Base2026.06 | 15.22 | |
| R-ZeroModel Backbone=Qwen3-4B-Base2026.06 | 14.57 | |
| BaseModel Backbone=Qwen3-8B-Base2026.06 | 14.49 | |
| AZRModel Backbone=Qwen3-8B-Base2026.06 | 14.49 | |
| INFUSERModel Backbone=Qwen3-4B-Base2026.06 | 13.78 | |
| BaseModel Backbone=Qwen3-4B-Base2026.06 | 13.02 | |
| AZRModel Backbone=Qwen3-4B-Base2026.06 | 12.65 |