| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Quality Rejection | Merged | Dice73.45 | 7 | |
| 3-Class Quality Assessment | Merged | QWK68.41 | 7 | |
| Violence Detection | Merged All-source (test) | Accuracy93.5 | 3 | |
| Sign Language Recognition | Merged-5565 | Top-1 Accuracy53.34 | 2 | |
| Vision-and-Language Navigation | Merged Unseen (test) | nDTW33.6 | 2 | |
| Vision-and-Language Navigation | Merged Unseen (dev) | nDTW36.3 | 2 | |
| Vision-and-Language Navigation | Merged Seen (test) | nDTW57.4 | 2 | |
| Vision-and-Language Navigation | Merged Seen (dev) | nDTW58.6 | 2 | |
| Frame-level Video Manipulation Detection | Merged authentic and manipulated frames (evaluation) | Global Accuracy83.12 | 1 |