| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Overall Evaluation | Demo-ICL-Bench | Average Score80.1 | 14 | |
| Demonstration Selection | Demo-ICL-Bench | Overall Accuracy76 | 14 | |
| Video-demo In-context Learning | Demo-ICL-Bench | Accuracy (Demo)80.4 | 14 | |
| Text-demo In-context Learning | Demo-ICL-Bench | Accuracy (Demo)84 | 14 | |
| Video In-Context Learning | Demo-ICL-Bench (test) | T ICL Score54.4 | 12 |