| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Short Video Question Answering | Vinoground | Text Count/Score65.8 | 12 | |
| Video Understanding | Vinoground video sub-task zero-shot | Accuracy (zero-shot)38.2 | 11 | |
| Video Understanding | Vinoground | Group Score20.2 | 10 | |
| General Video QA | vinoground | Accuracy47.6 | 4 |