通过解耦方向与幅度，实现视频生成的快速采样。
Magnitude-Direction Decoupling for Fast Video Generation with Flow Matching Models
- 将扩散模型的去噪过程分解为幅度和方向两部分，分别用轻量模型和缓存信息处理。
- 在Wan2.1数据集上实现最高2.95倍加速，同时保持高质量视觉效果。
- 适合追求高效视频生成且对延迟敏感的应用场景。
基于流匹配的视频生成模型虽性能出色，但因迭代去噪带来高计算开销。我们发现并非所有去噪步骤都需原模型，部分可采用轻量替代以提速。然而直接使用缓存或轻量模型会偏离原始去噪轨迹，影响效果。实证分析表明，轻量模型能稳健捕捉原模型输出的幅度分量，而缓存则提供可靠的方向引导。基于此，我们提出幅度-方向解耦（MDD）方法，自适应地以校准方向的轻量模型替代原模型，加速推理并有效修正轨迹偏差。此外，MDD在无分类器引导（CFG）下复用幅度信息，进一步降低推理成本。实验显示，MDD优于现有加速方法，在保持高视觉保真度与内容丰富性的同时，实现显著提速（如在Wan2.1上达2.95倍）。
原文摘要 · Abstract (English)
Flow matching models for video generation achieve impressive performance but suffer from high computational overhead due to iterative denoising. In fact, the original model is not necessary for all denoising steps, allowing some steps to use lightweight alternatives for faster sampling. However, directly using caching or lightweight models can deviate from the original denoising trajectory, resulting in suboptimal performance. Through empirical analysis, we find that lightweight models can robustly capture the magnitude components of the original model's output, while caching provides reliable directional guidance. Building on this insight, we propose the Magnitude-Direction Decoupling (MDD) method, which adaptively employs a direction-calibrated lightweight model as a substitute for the original model to accelerate inference and effectively correct deviations in the denoising trajectory. Moreover, MDD further reduces inference costs by reusing magnitude information under classifier-free guidance (CFG). As a result, MDD offers a more reliable and lightweight solution to accelerate sampling. Experiments show that MDD outperforms existing acceleration methods, delivering promising speedups (e.g., up to 2.95x on Wan2.1) while preserving high visual fidelity and content richness.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。