ST-Bench
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Spatiotemporal reasoning | ST-Bench 2026 | Exact Match (EM)55.3 | 8 | |
| Storyboard Generation | ST-Bench | BG Consistency (Consecutive)4.94 | 5 | |
| Multi-shot video storytelling | ST-Bench | Aesthetic Quality64.52 | 5 |
Showing 3 of 3 rows