| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Video Counting | Molmo2 VideoCount (val) | Accuracy37.1 | 43 | |
| Video Counting | Molmo2-VC | Accuracy37.1 | 24 | |
| Tracking | Molmo2 Track | Animals J&F0.81 | 17 | |
| Video pointing | Molmo2-VP | F1 Score39.9 | 13 | |
| Figure-to-SVG generation | Molmo2 Diagram | SSIM0.942 | 9 | |
| Rule-based evaluation of diagram components | Molmo | R60.8 | 5 |