Medical Instruction Following on MulDimIF
78.7Pass@1MedXIAOHE
Evaluation Results
| Method | Links | |
|---|---|---|
| MedXIAOHEMode=Thinking mode, Decoding=Greedy2026.02 | 78.7 | |
| GPT-5.2 ThinkingMode=Thinking mode, Decoding=Greedy2026.02 | 78.6 | |
| Gemini 3.0 ProDecoding=Greedy2026.02 | 77.8 | |
| Gemini 2.5 ProDecoding=Greedy2026.02 | 73.2 |