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.
@article{arxiv.2502.13129,
title = {Is Noise Conditioning Necessary for Denoising Generative Models?},
author = {Qiao Sun and Zhicheng Jiang and Hanhong Zhao and Kaiming He},
journal= {arXiv preprint arXiv:2502.13129},
year = {2025}
}
Comments
Update ImageNet experiments (SiT with CFG). Update Appendix