Interactive Video Generation (Smooth Transition) on VBench Long
29.77Text AlignmentLongLive + Echo-Forcing
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| LongLive + Echo-ForcingFine-tuning protocol=Fine-tuned2026.05 | 29.77 | 95.32 | 93.74 | 54.78 | 67.87 | |
| Echo-ForcingFine-tuning protocol=Unfine-tuned2026.05 | 27.94 | 93.19 | 92.63 | 58.64 | 69.21 | |
| LongLiveFine-tuning protocol=Fine-tuned2026.05 | 27.38 | 94.01 | 91.39 | 54.89 | 67.17 | |
| LongLive + FlushFine-tuning protocol=Fine-tuned2026.05 | 27.34 | 94.22 | 91.88 | 54.85 | 67.33 | |
| Self-Forcing + RecacheFine-tuning protocol=Unfine-tuned2026.05 | 26.92 | 86.13 | 86.48 | 50.15 | 50.22 | |
| ∞-RopeFine-tuning protocol=Unfine-tuned2026.05 | 25.94 | 83.95 | 87.94 | 57.33 | 66.81 |