| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Consistency evaluation suite N=720 (test) | Semantic Consistency Score0.583 | 54 | 3mo ago | ||
| Adult, Housing, Loan, and HELOC | Violation Rate (S1)2.24 | 10 | 1mo ago | ||
| UniGenBench In-domain v1 | FLUX.1-dev w/ UnifiedReward-Flex | Overall Score73.95 | 7 | 3mo ago | |
| WebVid10M | Collaborative Video Diffusion (Ours) | CLIP-F0.93 | 5 | 3mo ago | |
| ConceptRisk | VII | Kling CLIPp2c Score0.5052 | 3 | 3mo ago | |
| COCO-I2VSafetyBench | VII | Kling (CLIPp2c)0.5679 | 3 | 3mo ago |