Text-to-Video Generation on Subjective Evaluation 10 text prompts (test)
3.69User Preference ScoreAutoRefiner
Evaluation Results
| Method | Links | |
|---|---|---|
| AutoRefinerBase Model=Self-Forcing2025.12 | 3.69 | |
| Self-Forcing + InitNoiseRefinerRefiner=InitNoiseRefiner2025.12 | 2.31 | |
| Self-Forcing + LoRAAdaptation=LoRA2025.12 | 2.25 | |
| Self-Forcing2025.12 | 2.23 |