Medical Safety Evaluation on MedSafetyBench QR
100Safety ScoreRESTA
Evaluation Results
| Method | Links | |
|---|---|---|
| RESTABackbone Model=Lingshu-7B2025.12 | 100 | |
| PSIBackbone Model=HuatuoGPT-Vision-7B2025.12 | 98.22 | |
| RESTABackbone Model=HuatuoGPT-Vision-7B2025.12 | 98 | |
| PSIBackbone Model=Lingshu-7B2025.12 | 94.22 | |
| ModelMergeBackbone Model=Lingshu-7B2025.12 | 63.33 | |
| OriginalBackbone Model=Lingshu-7B2025.12 | 60 | |
| ModelMergeBackbone Model=HuatuoGPT-Vision-7B2025.12 | 50 | |
| OriginalBackbone Model=HuatuoGPT-Vision-7B2025.12 | 47.78 |