Medical Visual Question Answering on VQA_RAD (Pass@1)
74.5Pass@1Gemini 3.0 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 3.0 ProDecoding=Greedy2026.02 | 74.5 | |
| GPT-5.2 ThinkingMode=Thinking mode, Decoding=Greedy2026.02 | 68.96 | |
| Gemini 2.5 ProDecoding=Greedy2026.02 | 67.41 | |
| MedXIAOHEMode=Thinking mode, Decoding=Greedy2026.02 | 66.96 |