Is Noise Conditioning Necessary for Denoising Generative Models?

Qiao Sun, Zhicheng Jiang, Hanhong Zhao, Kaiming He
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:57469-57502, 2025.

Abstract

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 mathematical analysis of the error introduced 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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-sun25g, title = {Is Noise Conditioning Necessary for Denoising Generative Models?}, author = {Sun, Qiao and Jiang, Zhicheng and Zhao, Hanhong and He, Kaiming}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {57469--57502}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/sun25g/sun25g.pdf}, url = {https://proceedings.mlr.press/v267/sun25g.html}, abstract = {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 mathematical analysis of the error introduced 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.} }
Endnote
%0 Conference Paper %T Is Noise Conditioning Necessary for Denoising Generative Models? %A Qiao Sun %A Zhicheng Jiang %A Hanhong Zhao %A Kaiming He %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-sun25g %I PMLR %P 57469--57502 %U https://proceedings.mlr.press/v267/sun25g.html %V 267 %X 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 mathematical analysis of the error introduced 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.
APA
Sun, Q., Jiang, Z., Zhao, H. & He, K.. (2025). Is Noise Conditioning Necessary for Denoising Generative Models?. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:57469-57502 Available from https://proceedings.mlr.press/v267/sun25g.html.

Related Material