动量扩散采样以减轻发散伪影

摘要

尽管扩散模型在图像生成方面取得了显著成功，但缓慢的采样仍然是一个持续存在的问题。为加速采样过程，先前的研究已将扩散采样重新表述为ODE/SDE，并引入了高阶数值方法。然而，这些方法通常会产生发散伪影，特别是在采样步骤较少时，这限制了加速的实现。在本文中，我们调查了这些伪影的潜在原因，并提出这些方法稳定区域较小可能是主要原因。为解决这一问题，我们提出了两种新技术。第一种技术涉及将Heavy Ball（HB）动量，一种用于改善优化的众所周知技术，纳入现有的扩散数值方法以扩展它们的稳定区域。我们还证明了由此产生的方法具有一阶收敛性。第二种技术，称为广义Heavy Ball（GHVB），构建了一种新的高阶方法，提供了精度和伪影抑制之间的可变折衷。实验结果表明，我们的技术在减少伪影和提高图像质量方面非常有效，在像素级和潜在级扩散模型的低步采样上超越了最先进的扩散求解器。我们的研究为未来扩散工作的数值方法设计提供了新颖的见解。

English

Despite the remarkable success of diffusion models in image generation, slow sampling remains a persistent issue. To accelerate the sampling process, prior studies have reformulated diffusion sampling as an ODE/SDE and introduced higher-order numerical methods. However, these methods often produce divergence artifacts, especially with a low number of sampling steps, which limits the achievable acceleration. In this paper, we investigate the potential causes of these artifacts and suggest that the small stability regions of these methods could be the principal cause. To address this issue, we propose two novel techniques. The first technique involves the incorporation of Heavy Ball (HB) momentum, a well-known technique for improving optimization, into existing diffusion numerical methods to expand their stability regions. We also prove that the resulting methods have first-order convergence. The second technique, called Generalized Heavy Ball (GHVB), constructs a new high-order method that offers a variable trade-off between accuracy and artifact suppression. Experimental results show that our techniques are highly effective in reducing artifacts and improving image quality, surpassing state-of-the-art diffusion solvers on both pixel-based and latent-based diffusion models for low-step sampling. Our research provides novel insights into the design of numerical methods for future diffusion work.

动量扩散采样以减轻发散伪影

Diffusion Sampling with Momentum for Mitigating Divergence Artifacts

摘要

Support