| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| VideoRewardBench | VideoGRM | Perception (long)76.2 | 16 | 23d ago | |
| VideoGen-Bench | DeScore | Accuracy (w/ Tie)56.8 | 7 | 26d ago | |
| GenAI | UnifiedReward-Thinking | Accuracy (w/ Tie)54.8 | 7 | 26d ago | |
| In-domain Dataset | DeScore | Preference Accuracy73.4 | 7 | 26d ago |