挑战噪声条件必要性，发现无噪声条件模型仍能高效生成图像。
Is Noise Conditioning Necessary for Denoising Generative Models?
- 去掉噪声条件后，多数生成模型仍能稳定运行
- 提出无噪声条件模型，在CIFAR-10上达到FID 2.23
- 为生成模型设计提供新思路，适合关注基础机制的研究者
主流观点认为噪声条件对去噪扩散模型至关重要。本文基于盲图像去噪研究，探索无噪声条件下的多种去噪生成模型。令人意外的是，大多数模型表现出优雅退化，部分甚至性能更优。我们提供了移除噪声条件导致误差的理论分析，结果与实验一致。进一步提出一种无噪声条件模型，在CIFAR-10上实现2.23的FID，显著缩小与领先噪声条件模型的差距。希望本工作推动社区重新审视去噪生成模型的基础设定。
原文摘要 · Abstract (English)
It is widely believed that noise conditioning is indispensable for denoising diffusion models to work successfully. This work challenges this belief. Motivated by research on blind image denoising, we investigate a variety of denoising-based generative models in the absence of noise conditioning. To our surprise, most models exhibit graceful degradation, and in some cases, they even perform better without noise conditioning. We provide a theoretical analysis of the error caused by removing noise conditioning and demonstrate that our analysis aligns with empirical observations. We further introduce a noise-unconditional model that achieves a competitive FID of 2.23 on CIFAR-10, significantly narrowing the gap to leading noise-conditional models. We hope our findings will inspire the community to revisit the foundations and formulations of denoising generative models.
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。