Medical on MedQA
65.78AccuracyINFUSER
Evaluation Results
| Method | Links | |
|---|---|---|
| INFUSERModel Backbone=Qwen3-8B-Base2026.06 | 65.78 | |
| AZRModel Backbone=Qwen3-8B-Base2026.06 | 65.28 | |
| R-ZeroModel Backbone=Qwen3-8B-Base2026.06 | 65.12 | |
| BaseModel Backbone=Qwen3-8B-Base2026.06 | 64.18 | |
| AZRModel Backbone=Qwen3-4B-Base2026.06 | 59.62 | |
| R-ZeroModel Backbone=Qwen3-4B-Base2026.06 | 58.92 | |
| INFUSERModel Backbone=Qwen3-4B-Base2026.06 | 58.86 | |
| BaseModel Backbone=Qwen3-4B-Base2026.06 | 55.46 |