| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Physical Reasoning | PhyX | Accuracy48.6 | 24 | |
| Multimodal Autoformalization | PHYX Thermodynamics | Compile Rate20 | 10 | |
| Multimodal Autoformalization | PHYX Electromagnetism | Compile Success Rate0.667 | 10 | |
| Multimodal Autoformalization | PHYX Mechanics | Compile Rate71.4 | 10 | |
| Multimodal Autoformalization | PHYX Modern | Compile Rate71.4 | 10 | |
| Open-Ended | PhyX mini (test) | Overall Accuracy2.7 | 7 | |
| Multiple-Choice Question | PhyX mini (test) | Overall Accuracy46.2 | 7 |