用自适应动量稳定扩散采样中的噪声梯度，提升图像生成质量。
Adaptive Moments are Surprisingly Effective for Plug-and-Play Diffusion Sampling
- 用自适应动量估计降低扩散采样中梯度噪声。
- 在图像修复和类别条件生成上达到当前最优性能。
- 方法简单高效，适合对采样精度要求高的应用。
指导性扩散采样依赖于难以计算的似然梯度近似，这会引入显著噪声，影响采样动态。本文提出使用自适应动量估计来稳定采样过程中的噪声似然梯度。尽管方法简单，但在图像修复和类别条件生成任务上取得了当前最优结果，优于许多更复杂且计算开销更大的方法。我们在合成数据和真实数据上进行了实证分析，证明通过自适应动量抑制梯度噪声，能有效提升生成结果与目标分布的对齐程度。
原文摘要 · Abstract (English)
Guided diffusion sampling relies on approximating often intractable likelihood scores, which introduces significant noise into the sampling dynamics. We propose using adaptive moment estimation to stabilize these noisy likelihood scores during sampling. Despite its simplicity, our approach achieves state-of-the-art results on image restoration and class-conditional generation tasks, outperforming more complicated methods, which are often computationally more expensive. We provide empirical analysis of our method on both synthetic and real data, demonstrating that mitigating gradient noise through adaptive moments offers an effective way to improve alignment.
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。