arXiv：cs.LG（机器学习，全量分类）· Aneesh Barthakur, Mathias Niepert, Luiz F. O. Chamon·· 14 小时前AI 评分37
扩散模型的端到端学习引导调度（LEEGS）
Learned End-to-End Guidance Schedules for Diffusion Models
AI 导读
研究提出学习式端到端引导调度（LEEGS），通过在小样本集上用随机梯度下降训练时间相关调度，并用梯度近似将训练时间缩短至原来的1/4。在图像修复、噪声图像逆问题、Face-ID引导生成及PDE正逆问题等任务上，LEEGS在相同预算（50或100 NFEs）下优于基线，或以仅10%的步数达到恒定引导的效果。
正文
Abstract:Diffusion models are a powerful generative paradigm used across multimedia and scientific applications. Guided diffusion methods impose requirements on the generation by adding the gradient of a differentiable loss (the guidance function) as a drift term during inference. The weight of this drift (the guidance scale) is critical for the trade-off between data quality and requirement satisfaction. To achieve both of these goals, guided diffusion must resort to small guidance scales and lengthy sampling, incurring high computational costs. This work proposes learned end-to-end guidance schedules (LEEGS) to achieve these objectives with fewer sampling steps. LEEGS trains a time-dependent schedule by minimizing the guidance function over a small set of examples using stochastic gradient descent. Backpropagating through guided sampling is computationally expensive, so LEEGS uses an approximation of the gradient that cuts training time by a factor of 4. We evaluate LEEGS on diverse guidance tasks, including (a) image inpainting, (b) noisy image inverse problems, (c) face-ID-guided generation, and (d) forward and inverse PDE problems, outperforming baselines at equal budget (50 or 100 NFEs), or matching constant guidance with only 10% of the steps.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.01502 [cs.LG] |
| (or arXiv:2610.01502v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.01502 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Aneesh Barthakur [view email]
[v1]
Thu, 1 Oct 2026 11:42:48 UTC (32,396 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org