Metraipolis › archive › dc_d89093afcb5a

Rollout-Marginal Distillation for Long-Horizon Autoregressive Video Generation — arXiv:2609.37925

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.37925
License (as archived)
CC0-1.0
Archived
2026-10-07T16:27:55.451Z
SHA-256 of the archived text
cfb06b246d031e26e85bf5db5433eabc7c5253b8b1aaf8cce2ef8b5b5cf62612
Stable URL of this page
https://metraipolis.com/documento/dc_d89093afcb5a

How to cite

Rollout-Marginal Distillation for Long-Horizon Autoregressive Video Generation — arXiv:2609.37925 — arXiv (en), https://arxiv.org/abs/2609.37925 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d89093afcb5a (Metraipolis archive dc_d89093afcb5a, sha256 cfb06b246d031e26e85bf5db5433eabc7c5253b8b1aaf8cce2ef8b5b5cf62612)
BibTeX
@misc{dcd89093afcb5a,
  title = {Rollout-Marginal Distillation for Long-Horizon Autoregressive Video Generation — arXiv:2609.37925},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.37925},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d89093afcb5a; sha256 cfb06b246d031e26e85bf5db5433eabc7c5253b8b1aaf8cce2ef8b5b5cf62612},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37925. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autoregressive (AR) video diffusion enables low-latency, streamable video generation, but prediction errors often accumulate over long rollouts. Training the generator on its own rollouts exposes it to these imperfect histories. However, existing video-level distribution matching distillation (DMD) scores the whole rollout jointly. Because a chunk is evaluated together with its past and future, its correction can favor matching artifacts in the surrounding context merely to preserve temporal consistency. To provide a clearer visual-quality signal, we introduce Rollout-Marginal Distillation (RMD). RMD retains the generated history for AR prediction but scores each chunk independently against a chunk teacher, ensuring its quality correction is not compromised by an imperfect temporal context. To compensate for the lack of temporal context in independent chunk scoring, RMD subsequently applies video-level DMD to restore temporal coherence. Extensive experiments demonstrate that RMD maintains high visual quality far beyond its training horizon and outperforms video-level DMD baselines. Code and video results are available at this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.37925 — 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 2026-10-07T15:54:54.051Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).