从去噪视角揭示流匹配生成过程的动态阶段。
The Generation Phases of Flow Matching: a Denoising Perspective
- 将流匹配与去噪器统一在相同框架下对比
- 发现生成过程存在不同动态阶段，去噪效果随阶段变化
- 为改进生成质量提供可控制的扰动方法，适合生成模型研究者
流匹配已取得显著成功，但其生成过程的质量影响因素仍不明确。本文从去噪视角出发，构建框架以实证探究生成过程。通过建立流匹配模型与去噪器之间的形式化联系，为生成与去噪性能的比较提供共同基础。由此设计出有原则且可控的扰动（噪声与漂移），揭示生成过程的不同动态阶段，精确刻画去噪器在生成过程中何时成功或失败，及其重要性。
原文摘要 · Abstract (English)
Flow matching has achieved remarkable success, yet the factors influencing the quality of its generation process remain poorly understood. In this work, we adopt a denoising perspective and design a framework to empirically probe the generation process. Laying down the formal connections between flow matching models and denoisers, we provide a common ground to compare their performances on generation and denoising. This enables the design of principled and controlled perturbations to influence sample generation: noise and drift. This leads to new insights on the distinct dynamical phases of the generative process, enabling us to precisely characterize at which stage of the generative process denoisers succeed or fail and why this matters.
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。