扩散模型的校正训练：缩小训练-推理差距

AI 摘要

该论文针对扩散模型推理时的曝光偏差问题，提出一种校正训练方法：训练一个校正器以基础模型自身的预测为输入，并针对相同真实目标进行训练。在采样时，每个去噪步骤迭代精炼干净样本预测，可显著提升样本质量并保持多样性。方法在DiT、SiT、JiT等架构上验证，FID优于无引导生成，性能与CFG相当，且可通过蒸馏将校正能力回注基础模型，不增加推理开销。

正文节选

Correction Training for Diffusion Models Abstract Iterative sampling in diffusion models during inference can lead to error accumulation due to exposure bias. The denoiser is trained on real training images interpolated with noise, but at inference its inputs are constructed from its own imperfect predictions from the earlier steps, something it has never seen at training time. This might be one reason diffusion models often struggle without additional help, such as CFG. In this paper, we introd

发布时间：2026-10-15 12:00

抓取时间：2026-10-14 14:16

来源机构：arXiv