Signals 4 · free daily AI digest

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

Paper recorded by Signals 4 on 2026-10-02 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-02 on arXiv · recorded by Signals 4 on 2026-10-05

Category: cs.CV · 计算机视觉 · first seen 2026-10-05

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,

Read on arXiv →

#372 most recent of 756 cs.CV papers we have recorded · ↑ newer: ManifoldSplat: Language-Guided Semantic Shape Editing of 3D Gaussian H · ↓ older: Feedforward Novel View Synthesis for Heterogeneous Cameras
Cite this page: DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation: the #372 most recent of 756 cs.CV papers we have recorded (as of 2026-10-02). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/duomatching-joint-marginal-distribution-matching-for-few-step-video-generation.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CV papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions