解析生成模型训练中权重与参数化的关键影响，给出实用设计建议。
Training Flow Matching: The Role of Weighting and Parameterization
- 对比噪声、图像、速度三种参数化方式的训练表现
- 发现数据流形维度和数据量显著影响模型性能
- 适合关注生成模型训练机制的研究者和实践者
我们研究基于去噪的生成模型的训练目标，重点关注损失权重与输出参数化，包括基于噪声、干净图像和速度的公式。通过系统的数值实验，分析这些训练选择如何与数据流形的内在维度、模型架构及数据集规模相互作用。实验涵盖具有可控几何结构的合成数据集和真实图像数据，使用去噪精度（不同噪声水平下的PSNR）和生成质量（FID）作为定量指标。本文不提出新方法，而是旨在解耦训练流匹配模型时的关键影响因素，为实际设计提供可操作的洞见。
原文摘要 · Abstract (English)
We study the training objectives of denoising-based generative models, with a particular focus on loss weighting and output parameterization, including noise-, clean image-, and velocity-based formulations. Through a systematic numerical study, we analyze how these training choices interact with the intrinsic dimensionality of the data manifold, model architecture, and dataset size. Our experiments span synthetic datasets with controlled geometry as well as image data, and compare training objectives using quantitative metrics for denoising accuracy (PSNR across noise levels) and generative quality (FID). Rather than proposing a new method, our goal is to disentangle the various factors that matter when training a flow matching model, in order to provide practical insights on design choices.
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。