Metraipolis › archive › dc_b89cb77444fc

Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout — arXiv:2609.09123

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.09123
License (as archived)
CC0-1.0
Archived
2026-10-16T15:12:33.594Z
SHA-256 of the archived text
069d53f7ecb3f0d8b21e658e8ca9a4779d7e290f72b0ed4e5bae1624dcfa922d
Stable URL of this page
https://metraipolis.com/documento/dc_b89cb77444fc

How to cite

Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout — arXiv:2609.09123 — arXiv (en), https://arxiv.org/abs/2609.09123 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b89cb77444fc (Metraipolis archive dc_b89cb77444fc, sha256 069d53f7ecb3f0d8b21e658e8ca9a4779d7e290f72b0ed4e5bae1624dcfa922d)
BibTeX
@misc{dcb89cb77444fc,
  title = {Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout — arXiv:2609.09123},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.09123},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b89cb77444fc; sha256 069d53f7ecb3f0d8b21e658e8ca9a4779d7e290f72b0ed4e5bae1624dcfa922d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09123. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autoregressive (AR) video diffusion models have shown great potential in real-time video generation. Recent methods distill pretrained bidirectional video diffusion models into causal AR students through Distribution Matching Distillation (DMD), but the generated videos often suffer from over-saturation and over-smoothing issues, resulting in limited visual quality and realism. The key contributing factor is the mode-seeking behavior of the reverse KL objective in DMD, which can cause the student distribution to collapse onto only a few modes of the teacher distribution. To address this, we propose Mask Forcing, a Dual-Noise Masking Rollout strategy that perturbs the AR student self-rollout to mitigate mode collapse induced by reverse-KL mode seeking. The core idea is to inject cleaner signals into noisy rollout inputs via random masks along spatial and temporal axes during the self-rollout process of AR diffusion distillation. Such perturbations encourage the student rollouts to explore more regions of the teacher distribution, allowing DMD to provide learning signals beyond the modes already covered by the student. Moreover, the cleaner tokens act as denoising guidance for other noisier tokens, improving the intermediate rollout predictions and reducing error accumulation. Extensive experiments demonstrate that our method improves multiple AR video diffusion distillation methods with higher visual quality efficiently, without incorporating real video data or additional post-training stages.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.09123 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T14:49:57.394Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).