SOURCE-LINKED INTELLIGENCE

Uncertainty DMD: Restoring Diversity in Few-Step Autoregressive Video Distillation

arXiv · AI, language, vision and robotics · article · Sep 10, 2026 · UTC

Few-step distillation improves the efficiency of autoregressive (AR) video generation, but often causes diversity collapse: under the same prompt, different noise samples tend to produce highly similar videos with weakened motion dynamics. We analyze this degradation in Distribution Matching Distillation (DMD)-distilled AR video generators and find that, in the autoregressive setting, it takes the form of a structured uncertainty collapse: the mode-seeking bias of DMD maps different noise samples to nearly identical first chunks, and the deterministic AR cache then propagates this collapsed st

Read original source ↗ Open in workspace

recordType

paper

region

Global