arXiv：cs.LG· Shizheng Lin, Soon Hoe Lim, N. Benjamin Erichson·· 5 小时前AI 评分39
AREX：面向 Flow Matching 少步采样的仿射残差指数积分器
AREX: Affine-Residual Exponential Integrator for Few-Step Sampling in Flow Matching
AI 导读
AREX 是一种免训练的 flow matching 采样器，利用目标均值和协方差解析处理采样动态中的仿射部分，仅需对神经残差项做积分。该方法将学习到的动态分解为由目标前两阶矩确定的仿射分量与神经残差项，并用显式矩阵值传播子积分仿射分量，区别于仅能解析处理各向同性线性动态的标量指数积分器。在图像与文本到图像生成任务中，AREX 在少步采样场景下持续提升样本保真度，且无需重新训练底层模型。
正文
Abstract:We introduce AREX, a training-free sampler for pretrained flow matching models that uses the target mean and covariance to capture an analytically tractable part of the sampling dynamics. We show that the velocity field of the moment-matched Gaussian target is the $L^2$-optimal affine approximation to the marginal velocity field. This motivates decomposition of the learned dynamics into an affine component over the whole sampling path, determined by the first two target moments, and a neural residual term. AREX keeps the affine component and integrates it using an explicit matrix-valued propagator. In turn, we only require to integrate over the residual term. This differs from scalar exponential integrators, which analytically handle only isotropic linear dynamics. Across image and text-to-image generation tasks, AREX consistently improves sample fidelity in the few-step sampling regime without retraining the underlying model.
| Comments: | 53 pages |
| Subjects: | Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.03483 [stat.ML] |
| (or arXiv:2610.03483v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.03483 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Soon Hoe Lim [view email]
[v1]
Fri, 2 Oct 2026 15:50:49 UTC (13,089 KB)
来源：arXiv：cs.LG · arxiv.org