Megadose AI progress, ranked and analyzed.

DistillAlign: Coordinating Mode Covering and Mode Seeking in Autoregressive Video Distillation

· HF Daily Papers ·
DistillAlign argues that video distillation fails when the student learns sharp outputs but loses the teacher’s distributional coverage.

The paper says common DMD-based autoregressive video distillation splits initialization and refinement in ways that target different distributions. Its evaluation protocol compares student and teacher precision and coverage in a shared latent space, showing gaps that visual scores can miss. The proposed joint distillation pairs DMD’s mode-seeking loss with a consistency distillation constraint meant to preserve coverage and diversity. In experiments, it reports better quality, coverage, and diversity, including outperforming baselines refined with Wan-14B while using a Wan-1.3B DMD teacher. HF Daily Papers' note

score 4

Categories: Research