| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Medical Question Answering | NEJM | Accuracy73.13 | 16 | |
| Medical Diagnosis | NEJM | Rounds17.91 | 9 | |
| Medical Reasoning | NEJM (OOD) | Accuracy68.3 | 7 | |
| Multi-turn Clinical Simulation | NEJM Extension OOD (test) | Accuracy33.3 | 6 | |
| Multi-turn Clinical Simulation | NEJM OOD (test) | Accuracy46.7 | 6 |