MSBench
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Single-frame Image Consistency | MSBench Single-subject v1 | CLIP Image Consistency0.872 | 11 | |
| Visual Narrative Generation | MSBench Single-subject short-frame visual narratives | CLIP-I92.8 | 9 | |
| Multi-subject image consistency | MSBench multi-subject | CLIP-I0.749 | 7 | |
| Single-frame Image Consistency | MSBench Multi-subject v1 | CLIP-I0.749 | 7 | |
| Visual Narrative Generation | MSBench Multi-subject short-frame visual narratives | CLIP-I0.765 | 6 |
Showing 5 of 5 rows