| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Joint Optic Cup and Optic Disc Segmentation | BASE2 | DOD95.73 | 17 | |
| Species identification | BASE dataset 2019A2 (test) | Balanced Accuracy97.7 | 12 | |
| LLM Steering | BASE→USER.POS (out-of-distribution) | Steerability2.535 | 8 | |
| Multi-Robot Task Allocation | base baseline scenario R20 V100 | Victims Rescued32,807 | 6 | |
| Damage Assessment (Damage Classes) | BASE dataset 2019A2 (test) | Balanced Accuracy92.2 | 2 |