Text-to-Video Generation on T2V-CompBench (test)
0.2755CLIPScoreBaseline
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| BaselineModel=Wan2.1-14B, prompt_type=negative-prompted CFG2026.03 | 0.2755 | 0.2147 | 23.1646 | 0.877 | 1.7147 | 0.0378 | 0.2561 | 0.1995 | |
| CADSModel=Wan2.1-1.3B2026.03 | 0.2654 | 0.2127 | 24.1668 | 0.7018 | 1.7996 | 0.0348 | 0.3536 | 0.2022 | |
| BaselineModel=Wan2.1-1.3B, prompt_type=negative-prompted CFG2026.03 | 0.2651 | 0.2108 | 22.7719 | 0.772 | 3.0671 | 0.0463 | 0.3517 | 0.2528 | |
| CADSModel=Wan2.1-14B2026.03 | 0.263 | 0.2121 | 23.1093 | 0.7853 | 3.5325 | 0.0438 | 0.3464 | 0.2286 | |
| MotionCFGModel=Wan2.1-14B2026.03 | 0.2593 | 0.2097 | 23.0819 | 0.8025 | 4.4839 | 0.0727 | 0.4213 | 0.3057 | |
| IGModel=Wan2.1-14B2026.03 | 0.2579 | 0.2092 | 23.227 | 0.8223 | 1.7266 | 0.0571 | 0.3893 | 0.2581 | |
| IGModel=Wan2.1-1.3B2026.03 | 0.2552 | 0.209 | 23.9607 | 0.6785 | 4.1116 | 0.0601 | 0.4761 | 0.3077 | |
| MotionCFGModel=Wan2.1-1.3B2026.03 | 0.255 | 0.2085 | 23.708 | 0.6378 | 6.5136 | 0.0726 | 0.5034 | 0.3531 |