跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-02AI 評分36

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

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

AI 導讀

DuoMatching 提出一種用於少步影片生成的分佈匹配框架,旨在改善視覺質素、構圖及語義對齊,同時大致保留動態效果。框架加入來自圖像生成器的逐影格監督,並以 LatentBridge 處理影片學生模型與圖像教師模型的潛在表示差異,以 Latent Variation Sampling 將監督分配至不同時間片段。人類評估中,DuoMatching 相對所有受評基線的整體偏好率均高於 80%。

正文

Abstract

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 https://johnzhan2023.github.io/DuoMatching/.

View arXiv page View PDF Project page GitHub 4 Add to collection

Get this paper in your agent:

hf papers read 2610.03543

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.03543 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.03543 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.03543 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co