通过高斯化预处理提升扩散模型生成质量，尤其改善小网络早期采样效果。
Enhancing diffusion models with Gaussianization preprocessing
- 对训练数据进行高斯化预处理，使目标分布更接近独立高斯分布。
- 小规模网络下早期生成质量显著提升，采样过程更稳定高效。
- 方法通用性强，适用于多种生成任务，无需修改模型结构。
扩散模型在图像生成等任务中表现出色，但其采样速度慢的问题源于轨迹分叉前延迟，导致早期重建质量较差。本文提出对训练数据进行高斯化预处理，使目标分布更接近独立高斯分布，从而简化模型学习任务，提升重建初期的生成质量，尤其对小规模网络效果显著。该方法不依赖模型结构，原则上可推广至各类生成任务，实现更稳定高效的采样过程。
原文摘要 · Abstract (English)
Diffusion models are a class of generative models that have demonstrated remarkable success in tasks such as image generation. However, one of the bottlenecks of these models is slow sampling due to the delay before the onset of trajectory bifurcation, at which point substantial reconstruction begins. This issue degrades generation quality, especially in the early stages. Our primary objective is to mitigate bifurcation-related issues by preprocessing the training data to enhance reconstruction quality, particularly for small-scale network architectures. Specifically, we propose applying Gaussianization preprocessing to the training data to make the target distribution more closely resemble an independent Gaussian distribution, which serves as the initial density of the reconstruction process. This preprocessing step simplifies the model's task of learning the target distribution, thereby improving generation quality even in the early stages of reconstruction with small networks. The proposed method is, in principle, applicable to a broad range of generative tasks, enabling more stable and efficient sampling processes.
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。