arXiv cs.AI· Yuchen Jiao, Na Li, Changxiao Cai, Yuting Wei, Gen Li·· 10 小时前AI 评分31
前向值离散化：无需高阶近似的更快随机扩散采样
Forward-Value Discretization: Faster Stochastic Diffusion Sampling Without High-Order Approximation
AI 导读
研究证明随机扩散采样器即使采用高阶近似仍保持一阶收敛，因此提出免训练的前向值离散化框架，用历史预测构造前瞻估计来评估模型。确定性变体 F-DPMSolver 在低 NFE 预算下提升样本质量，随机变体 F-DDPM 在 CIFAR-10、ImageNet、FFHQ 和 LSUN 上于相同 NFE 预算下取得更低 FID，仅需确定性基线 1.5 倍 NFE 即可达到有竞争力的样本质量。
正文 · 原文
Abstract:Stochastic diffusion samplers often require substantially more neural function evaluations (NFEs) than deterministic samplers to achieve comparable sample quality. Although higher-order discretization accelerates deterministic sampling, its benefits for stochastic sampling remains unclear. We establish matching upper and worst-case lower bounds showing that stochastic samplers with higher-order approximations remain first-order convergent, even with exact score functions. Thus, increasing the approximation order does not generally improve the convergence order. Motivated by this limitation, we introduce forward-value discretization, a training-free framework that evaluates the data/noise prediction model at a lookahead estimate constructed from previous predictions. For deterministic sampling, the resulting Forward DPMSolver (F-DPMSolver), improves sample quality at low NFE budgets through its DDIM-based variant, despite remaining first-order convergent. For stochastic sampling, we develop Forward DDPM (F-DDPM), which evaluates the model at a cleaner lookahead point adjusted for subsequent noise injection. This estimate reuses both the previous model prediction and the previously injected noise. Both variants admit first-order convergence guarantees with one model evaluation per iteration. Across CIFAR-10, ImageNet, FFHQ, and LSUN, F-DDPM consistently achieves lower FID than the evaluated stochastic baselines at matched NFE budgets. With 1.5 times as many NFEs as deterministic baselines, it achieves competitive sample quality, narrowing the practical efficiency gap between stochastic and deterministic sampling. Code is available at this https URL.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML) |
| Cite as: | arXiv:2611.14328 [cs.LG] |
| (or arXiv:2611.14328v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.14328 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuchen Jiao [view email]
[v1]
Tue, 10 Nov 2026 16:53:27 UTC (6,402 KB)
来源：arXiv cs.AI · arxiv.org