| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| CaseHOLD | No Defense | Agreement (%)41.5 | 8 | 1mo ago | |
| ConvFinQA | No Defense | Agreement (%)37.8 | 8 | 1mo ago | |
| FinQA | Output Pert. | Agreement (%)33.7 | 8 | 1mo ago | |
| MedMCQA | No Defense | Agreement (%)50.2 | 8 | 1mo ago | |
| MedQA | No Defense | Agreement62.3 | 8 | 1mo ago |