Medical Question Answering on SuperGPQA
27.66AccuracyOriginal
Evaluation Results
| Method | Links | |
|---|---|---|
| OriginalBackbone Model=Lingshu-7B2025.12 | 27.66 | |
| ModelMergeBackbone Model=Lingshu-7B2025.12 | 27.59 | |
| PSIBackbone Model=Lingshu-7B2025.12 | 27.44 | |
| ModelMergeBackbone Model=HuatuoGPT-Vision-7B2025.12 | 25.52 | |
| OriginalBackbone Model=HuatuoGPT-Vision-7B2025.12 | 22.49 | |
| PSIBackbone Model=HuatuoGPT-Vision-7B2025.12 | 22.29 | |
| RESTABackbone Model=HuatuoGPT-Vision-7B2025.12 | 21.42 | |
| RESTABackbone Model=Lingshu-7B2025.12 | 19.56 |