| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Lesion Segmentation | BT | mDice84.5 | 18 | |
| Expected Calibration Error | BT | ECE (15-bin)0.025 | 9 | |
| Preference Modeling | BT | NLL1.08 | 8 | |
| Prompt Optimization | BT | Best Score62 | 8 | |
| Multiple choice question answering | BT (test) | Accuracy39.6 | 7 | |
| Classification | BT UCI (train) | Accuracy79.59 | 6 |