Elucidating the SNR-t Bias of Diffusion Probabilistic Models

• URL: http://arxiv.org/abs/2604.16044v1
• Date: Fri, 17 Apr 2026 13:16:15 GMT
• Title: Elucidating the SNR-t Bias of Diffusion Probabilistic Models
• Abstract summary: Diffusion models often suffer from a Signal-to-Noise Ratio-timestep (SNR-t) bias.<n>We propose a simple yet effective differential correction method to mitigate the SNR-t bias.<n>Our approach significantly improves the generation quality of various diffusion models.
• Score: 28.312988944419516
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion Probabilistic Models have demonstrated remarkable performance across a wide range of generative tasks. However, we have observed that these models often suffer from a Signal-to-Noise Ratio-timestep (SNR-t) bias. This bias refers to the misalignment between the SNR of the denoising sample and its corresponding timestep during the inference phase. Specifically, during training, the SNR of a sample is strictly coupled with its timestep. However, this correspondence is disrupted during inference, leading to error accumulation and impairing the generation quality. We provide comprehensive empirical evidence and theoretical analysis to substantiate this phenomenon and propose a simple yet effective differential correction method to mitigate the SNR-t bias. Recognizing that diffusion models typically reconstruct low-frequency components before focusing on high-frequency details during the reverse denoising process, we decompose samples into various frequency components and apply differential correction to each component individually. Extensive experiments show that our approach significantly improves the generation quality of various diffusion models (IDDPM, ADM, DDIM, A-DPM, EA-DPM, EDM, PFGM++, and FLUX) on datasets of various resolutions with negligible computational overhead. The code is at https://github.com/AMAP-ML/DCW.

Related papers

• Improved denoising diffusion probabilistic models with efficient non-diagonal covariance modeling [33.785721262421]
  We develop a novel covariance model that captures the peculiar statistical structure of natural images.<n>We show improved performance compared to previous SOTA denoising samplers.
  arXiv  Detail & Related papers  (2026-08-22T14:21:55Z)
• Mitigating the Noise Shift for Denoising Generative Models via Noise Awareness Guidance [54.88271057438763]
  Noise Awareness Guidance (NAG) is a correction method that explicitly steers sampling trajectories to remain consistent with the pre-defined noise schedule.<n>NAG consistently mitigates noise shift and substantially improves the generation quality of mainstream diffusion models.
  arXiv  Detail & Related papers  (2025-10-14T13:31:34Z)
• Noise Conditional Variational Score Distillation [60.38982038894823]
  Noise Conditional Variational Score Distillation (NCVSD) is a novel method for distilling pretrained diffusion models into generative denoisers.<n>By integrating this insight into the Variational Score Distillation framework, we enable scalable learning of generative denoisers.
  arXiv  Detail & Related papers  (2025-06-11T06:01:39Z)
• A Fourier Space Perspective on Diffusion Models [6.834230686279937]
  Diffusion models are state-of-the-art generative models on data modalities such as images, audio, proteins and materials.<n>We study the inductive bias of the forward process of diffusion models in Fourier space.
  arXiv  Detail & Related papers  (2025-05-16T14:13:02Z)
• Non-stationary Diffusion For Probabilistic Time Series Forecasting [3.7687375904925484]
  We develop a diffusion-based probabilistic forecasting framework, termed Non-stationary Diffusion (NsDiff)<n>NsDiff combines a denoising diffusion-based conditional generative model with a pre-trained conditional mean and variance estimator.<n>Experiments conducted on nine real-world and synthetic datasets demonstrate the superior performance of NsDiff compared to existing approaches.
  arXiv  Detail & Related papers  (2025-05-07T09:29:39Z)
• Boosting Diffusion Models with Moving Average Sampling in Frequency Domain [101.43824674873508]
  Diffusion models rely on the current sample to denoise the next one, possibly resulting in denoising instability. In this paper, we reinterpret the iterative denoising process as model optimization and leverage a moving average mechanism to ensemble all the prior samples. We name the complete approach "Moving Average Sampling in Frequency domain (MASF)"
  arXiv  Detail & Related papers  (2024-03-26T16:57:55Z)
• One More Step: A Versatile Plug-and-Play Module for Rectifying Diffusion Schedule Flaws and Enhancing Low-Frequency Controls [77.42510898755037]
  One More Step (OMS) is a compact network that incorporates an additional simple yet effective step during inference. OMS elevates image fidelity and harmonizes the dichotomy between training and inference, while preserving original model parameters. Once trained, various pre-trained diffusion models with the same latent domain can share the same OMS module.
  arXiv  Detail & Related papers  (2023-11-27T12:02:42Z)
• Soft Mixture Denoising: Beyond the Expressive Bottleneck of Diffusion Models [76.46246743508651]
  We show that current diffusion models actually have an expressive bottleneck in backward denoising. We introduce soft mixture denoising (SMD), an expressive and efficient model for backward denoising.
  arXiv  Detail & Related papers  (2023-09-25T12:03:32Z)
• PartDiff: Image Super-resolution with Partial Diffusion Models [3.8435187580887717]
  Denoising diffusion probabilistic models (DDPMs) have achieved impressive performance on various image generation tasks. DDPMs generate new data by iteratively denoising from random noise. But diffusion-based generative models suffer from high computational costs due to the large number of denoising steps. This paper proposes the Partial Diffusion Model (PartDiff), which diffuses the image to an intermediate latent state instead of pure random noise.
  arXiv  Detail & Related papers  (2023-07-21T22:11:23Z)
• Semi-Implicit Denoising Diffusion Models (SIDDMs) [50.30163684539586]
  Existing models such as Denoising Diffusion Probabilistic Models (DDPM) deliver high-quality, diverse samples but are slowed by an inherently high number of iterative steps. We introduce a novel approach that tackles the problem by matching implicit and explicit factors. We demonstrate that our proposed method obtains comparable generative performance to diffusion-based models and vastly superior results to models with a small number of sampling steps.
  arXiv  Detail & Related papers  (2023-06-21T18:49:22Z)
• On Diffusion Modeling for Anomaly Detection [13.312007032203857]
  Diffusion models are attractive candidates for density-based anomaly detection.<n>We show that diffusion-based anomaly detection methods perform competitively for both semi-supervised and unsupervised settings.<n>These results establish diffusion-based anomaly detection as a scalable alternative to traditional methods.
  arXiv  Detail & Related papers  (2023-05-29T20:19:45Z)
• Denoising Diffusion Samplers [41.796349001299156]
  Denoising diffusion models are a popular class of generative models providing state-of-the-art results in many domains. We explore a similar idea to sample approximately from unnormalized probability density functions and estimate their normalizing constants. While score matching is not applicable in this context, we can leverage many of the ideas introduced in generative modeling for Monte Carlo sampling.
  arXiv  Detail & Related papers  (2023-02-27T14:37:16Z)
• Diffusion Model Based Posterior Sampling for Noisy Linear Inverse Problems [14.809545109705256]
  This paper presents a fast and effective solution by proposing a simple closed-form approximation to the likelihood score. For both diffusion and flow-based models, extensive experiments are conducted on various noisy linear inverse problems. Our method demonstrates highly competitive or even better reconstruction performances while being significantly faster than all the baseline methods.
  arXiv  Detail & Related papers  (2022-11-20T01:09:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.