Frequency Regulation for Exposure Bias Mitigation in Diffusion Models

• URL: http://arxiv.org/abs/2507.10072v2
• Date: Mon, 04 Aug 2025 12:53:13 GMT
• Title: Frequency Regulation for Exposure Bias Mitigation in Diffusion Models
• Authors: Meng Yu, Kun Zhan,
• Abstract summary: We make a key observation: the energy of predicted noisy samples in the reverse process continuously declines compared to perturbed samples in the forward process.<n>We introduce a dynamic frequency regulation mechanism utilizing wavelet transforms, which separately adjusts the low- and high-frequency subbands.<n>We derive the rigorous mathematical form of exposure bias.
• Score: 13.095683155232281
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion models exhibit impressive generative capabilities but are significantly impacted by exposure bias. In this paper, we make a key observation: the energy of predicted noisy samples in the reverse process continuously declines compared to perturbed samples in the forward process. Building on this, we identify two important findings: 1) The reduction in energy follows distinct patterns in the low-frequency and high-frequency subbands; 2) The subband energy of reverse-process reconstructed samples is consistently lower than that of forward-process ones, and both are lower than the original data samples. Based on the first finding, we introduce a dynamic frequency regulation mechanism utilizing wavelet transforms, which separately adjusts the low- and high-frequency subbands. Leveraging the second insight, we derive the rigorous mathematical form of exposure bias. It is worth noting that, our method is training-free and plug-and-play, significantly improving the generative quality of various diffusion models and frameworks with negligible computational cost. The source code is available at https://github.com/kunzhan/wpp.

Related papers

• How I Met Your Bias: Investigating Bias Amplification in Diffusion Models [11.80771961784834]
  Diffusion-based generative models demonstrate state-of-the-art performance across various image synthesis tasks.<n>Previous research has viewed bias amplification as an inherent characteristic of diffusion models.<n>We empirically demonstrate that samplers for diffusion models have a significant and measurable effect on bias amplification.
  arXiv  Detail & Related papers  (2025-12-23T10:46:48Z)
• Warm Diffusion: Recipe for Blur-Noise Mixture Diffusion Models [14.963168108932761]
  We argue that hot diffusion fails to exploit the strong correlation between high-frequency image detail and low-frequency structures.<n>We propose a unified Blur-Noise Mixture Diffusion Model (BNMD) to control blurring and noise jointly.
  arXiv  Detail & Related papers  (2025-11-21T02:37:09Z)
• Frequency Domain-Based Diffusion Model for Unpaired Image Dehazing [92.61216319417208]
  We propose a novel frequency domain-based diffusion model, named ours, for fully exploiting the beneficial knowledge in unpaired clear data.<n>Inspired by the strong generative ability shown by Diffusion Models (DMs), we tackle the dehazing task from the perspective of frequency domain reconstruction.
  arXiv  Detail & Related papers  (2025-07-02T01:22:46Z)
• 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)
• ReDDiT: Rehashing Noise for Discrete Visual Generation [41.72349583047408]
  We propose the rehashing noise framework for discrete diffusion transformer, termed ReDDiT, to extend absorbing states and improve expressive capacity of discrete diffusion models.<n>Experiments show that ReDDiT significantly outperforms the baseline (reducing gFID from 6.18 to 1.61) and is on par with the continuous counterparts with higher efficiency.
  arXiv  Detail & Related papers  (2025-05-26T08:17:20Z)
• 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)
• Denoising Score Distillation: From Noisy Diffusion Pretraining to One-Step High-Quality Generation [82.39763984380625]
  We introduce denoising score distillation (DSD), a surprisingly effective and novel approach for training high-quality generative models from low-quality data.<n>DSD pretrains a diffusion model exclusively on noisy, corrupted samples and then distills it into a one-step generator capable of producing refined, clean outputs.
  arXiv  Detail & Related papers  (2025-03-10T17:44:46Z)
• Shaping Inductive Bias in Diffusion Models through Frequency-Based Noise Control [43.87692887705523]
  We build inductive biases into the training and sampling of Diffusion Probabilistic Models (DPMs)<n>We show that different datasets necessitate different inductive biases, and that appropriate frequency-based noise control induces increased generative performance compared to standard diffusion.
  arXiv  Detail & Related papers  (2025-02-14T15:46:37Z)
• 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)
• Frequency Compensated Diffusion Model for Real-scene Dehazing [6.105813272271171]
  We consider a dehazing framework based on conditional diffusion models for improved generalization to real haze. The proposed dehazing diffusion model significantly outperforms state-of-the-art methods on real-world images.
  arXiv  Detail & Related papers  (2023-08-21T06:50:44Z)
• Low-Light Image Enhancement with Wavelet-based Diffusion Models [50.632343822790006]
  Diffusion models have achieved promising results in image restoration tasks, yet suffer from time-consuming, excessive computational resource consumption, and unstable restoration. We propose a robust and efficient Diffusion-based Low-Light image enhancement approach, dubbed DiffLL.
  arXiv  Detail & Related papers  (2023-06-01T03:08:28Z)
• DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection [80.20339155618612]
  DiffusionAD is a novel anomaly detection pipeline comprising a reconstruction sub-network and a segmentation sub-network.<n>A rapid one-step denoising paradigm achieves hundreds of times acceleration while preserving comparable reconstruction quality.<n>Considering the diversity in the manifestation of anomalies, we propose a norm-guided paradigm to integrate the benefits of multiple noise scales.
  arXiv  Detail & Related papers  (2023-03-15T16:14:06Z)
• DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration [103.79030498369319]
  Self-supervised diffusion model for hyperspectral image restoration is proposed. textttDDS2M enjoys stronger ability to generalization compared to existing diffusion-based methods. Experiments on HSI denoising, noisy HSI completion and super-resolution on a variety of HSIs demonstrate textttDDS2M's superiority over the existing task-specific state-of-the-arts.
  arXiv  Detail & Related papers  (2023-03-12T14:57:04Z)
• SinDiffusion: Learning a Diffusion Model from a Single Natural Image [159.4285444680301]
  We present SinDiffusion, leveraging denoising diffusion models to capture internal distribution of patches from a single natural image. It is based on two core designs. First, SinDiffusion is trained with a single model at a single scale instead of multiple models with progressive growing of scales. Second, we identify that a patch-level receptive field of the diffusion network is crucial and effective for capturing the image's patch statistics.
  arXiv  Detail & Related papers  (2022-11-22T18:00:03Z)
• How Much is Enough? A Study on Diffusion Times in Score-based Generative Models [76.76860707897413]
  Current best practice advocates for a large T to ensure that the forward dynamics brings the diffusion sufficiently close to a known and simple noise distribution. We show how an auxiliary model can be used to bridge the gap between the ideal and the simulated forward dynamics, followed by a standard reverse diffusion process.
  arXiv  Detail & Related papers  (2022-06-10T15:09:46Z)
• Diffusion-GAN: Training GANs with Diffusion [135.24433011977874]
  Generative adversarial networks (GANs) are challenging to train stably. We propose Diffusion-GAN, a novel GAN framework that leverages a forward diffusion chain to generate instance noise. We show that Diffusion-GAN can produce more realistic images with higher stability and data efficiency than state-of-the-art GANs.
  arXiv  Detail & Related papers  (2022-06-05T20:45:01Z)
• On Analyzing Generative and Denoising Capabilities of Diffusion-based Deep Generative Models [18.018935233383935]
  Diffusion-based Deep Generative Models (DDGMs) offer state-of-the-art performance in generative modeling. We study how the small amount of noise is transformed during the backward diffusion process.
  arXiv  Detail & Related papers  (2022-05-31T19:29:27Z)
• Come-Closer-Diffuse-Faster: Accelerating Conditional Diffusion Models for Inverse Problems through Stochastic Contraction [31.61199061999173]
  Diffusion models have a critical downside - they are inherently slow to sample from, needing few thousand steps of iteration to generate images from pure Gaussian noise. We show that starting from Gaussian noise is unnecessary. Instead, starting from a single forward diffusion with better initialization significantly reduces the number of sampling steps in the reverse conditional diffusion. New sampling strategy, dubbed ComeCloser-DiffuseFaster (CCDF), also reveals a new insight on how the existing feedforward neural network approaches for inverse problems can be synergistically combined with the diffusion models.
  arXiv  Detail & Related papers  (2021-12-09T04:28:41Z)

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.