Metraipolis › archive › dc_55f65dacfef2

DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation — arXiv:2610.03543

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.03543
License (as archived)
CC0-1.0
Archived
2026-10-06T09:08:49.972Z
SHA-256 of the archived text
539d63f24fd2fc9df3dcaeb3770bb5f09a89b1a2f4800a552fef3e0ff4936469
Stable URL of this page
https://metraipolis.com/documento/dc_55f65dacfef2

How to cite

DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation — arXiv:2610.03543 — arXiv (en), https://arxiv.org/abs/2610.03543 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_55f65dacfef2 (Metraipolis archive dc_55f65dacfef2, sha256 539d63f24fd2fc9df3dcaeb3770bb5f09a89b1a2f4800a552fef3e0ff4936469)
BibTeX
@misc{dc55f65dacfef2,
  title = {DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation — arXiv:2610.03543},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.03543},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_55f65dacfef2; sha256 539d63f24fd2fc9df3dcaeb3770bb5f09a89b1a2f4800a552fef3e0ff4936469},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.03543. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Streaming video generation has benefited from distribution matching distillation (DMD), which matches the joint distribution of video frames to a video teacher's approximation of the real video distribution. Although this joint matching mitigates drift during autoregressive rollouts, limitations remain in visual quality and semantic alignment. To address these limitations, we propose DuoMatching, a distribution matching framework that approximates the real video distribution through a unified joint-marginal formulation. On top of existing joint matching formulations, the additional marginal matching objective provides dedicated frame-level supervision from an image generator, transferring complementary visual and semantic priors from it. To apply this frame-level supervision in video generation, we introduce LatentBridge to resolve the latent representation mismatch between the video student and the image teacher. Latent Variation Sampling further distributes such frame-level supervision across distinct temporal segments, reducing redundancy. Experiments demonstrate that DuoMatching improves visual quality, composition, and semantic alignment while largely preserving motion dynamics. Human evaluations show overall preference rates above 80% against all evaluated baselines. The project page is available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-05 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.03543 — 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 2026-10-06T08:21:44.263Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).