Medical Question Answering on MedQA MCMLE
83.95AccuracyModelMerge
Evaluation Results
| Method | Links | |
|---|---|---|
| ModelMergeBackbone Model=HuatuoGPT-Vision-7B2025.12 | 83.95 | |
| ModelMergeBackbone Model=Lingshu-7B2025.12 | 75.98 | |
| OriginalBackbone Model=Lingshu-7B2025.12 | 75.13 | |
| PSIBackbone Model=Lingshu-7B2025.12 | 74.56 | |
| OriginalBackbone Model=HuatuoGPT-Vision-7B2025.12 | 73.53 | |
| PSIBackbone Model=HuatuoGPT-Vision-7B2025.12 | 70.37 | |
| RESTABackbone Model=HuatuoGPT-Vision-7B2025.12 | 61.3 | |
| RESTABackbone Model=Lingshu-7B2025.12 | 32.11 |