cs.CVOct 20, 2026

PEEK: One-Step-Look-Ahead Exposure Bias Correction for Diffusion Sampling

Authors: Yunwei Bai, Ying Kiat Tan, Tsuhan Chen

Organizations: National University of Singapore

Abstract

Diffusion image generators owe much of their practical success to accurate denoising networks and strong numerical samplers. However, during sampling, the denoiser predicts the next step from flawed inputs that differ substantially from real data. This train-test gap, or exposure bias, caps image generation quality (i.e., blurry images, missing details, or distorted shapes). To mitigate the problem, we propose PEEK, a training-free deterministic sampler plugin. At each eligible denoising step, PEEK first performs a provisional one-step update, evaluates the denoiser once at the look-ahead state, and interpolates the current clean prediction with this one-step-later prediction to form the next noisy state. By taking the additional step, we obtain a generally more image like prediction from the higher-signal noisy state, mitigating exposure bias. To optimize the interpolation positions, PEEK searches for an interpolation schedule with a greedy metric-driven search made efficient by the PEEK prefix and baseline noise-schedule-rebasing caches. Across pixel-space and latent-space diffusion models with unconditional, label-guided, and text-guided settings on CIFAR-10, LSUNBedroom,I mageNet, and MS-COCO,the look-ahead correction consistently improves over baselines matched by number of function evaluations (NFE), with Fréchet Inception Distance (FID) reductions up to 49.3%. The one-time per-model calibration, the only extra cost, takes at most 2.2 minutes in our experiments. Project website is available at https://dachengruoque.github.io/PEEK-Video/.

Figures & tables

Appendix figures & tables15 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Unlocking Few-Step Diffusion for Faithful Previews

    Sep 28, 2026Jing Jia, Sifan Liu, Guanyang WangDiffusion SamplingFew-Step Diffusion Sampling

  2. High-Quality Exposure Correction with Diffusion-Based Image Generation Priors

    Aug 9, 2026Ziwen Li, Meng Cao, Jinpu Zhang +4Image EnhancementDiffusion Model Fine-Tuning

  3. MixFlow Training: Alleviating Exposure Bias with Slowed Interpolation Mixture

    Dec 22, 2025Hui Li, Fu-Yun Wang, Haoyuan Xia +4Diffusion Model Fine-TuningDiffusion-Based Image Generation