arXiv：cs.LG（机器学习，全量分类）· Alexander Denker, Zeljko Kereta, Carola-Bibiane Sch\"onlieb, Moshe Eliasof·· 17 小时前AI 评分32
MS-Flow：用轨迹拼接求解基于流模型的逆问题
Trajectory Stitching for Solving Inverse Problems with Flow-Based Models
AI 导读
MS-Flow 将生成轨迹表示为一系列中间隐状态而非单一初始噪声，通过局部约束流动力学并用轨迹匹配惩罚耦合各段，交替更新中间隐状态与观测数据一致性，从而降低内存消耗并提升重建质量。该方法在图像修复、超分辨率和计算机断层扫描等逆问题上优于现有方法。
正文
Abstract:Flow-based generative models have emerged as powerful priors for solving inverse problems. One option is to directly optimize the initial latent code (noise), such that the flow output solves the inverse problem. However, this requires backpropagating through the entire generative trajectory, incurring high memory costs and numerical instability. We propose MS-Flow, which represents the trajectory as a sequence of intermediate latent states rather than a single initial code. By enforcing the flow dynamics locally and coupling segments through trajectory-matching penalties, MS-Flow alternates between updating intermediate latent states and enforcing consistency with observed data. This reduces memory consumption while improving reconstruction quality. We demonstrate the effectiveness of MS-Flow over existing methods on image recovery and inverse problems, including inpainting, super-resolution, and computed tomography.
| Subjects: | Image and Video Processing (eess.IV); Machine Learning (cs.LG) |
| MSC classes: | 68T07 |
| Cite as: | arXiv:2602.08538 [eess.IV] |
| (or arXiv:2602.08538v2 [eess.IV] for this version) | |
| https://doi.org/10.48550/arXiv.2602.08538 arXiv-issued DOI via DataCite |
Submission history
From: Alexander Denker [view email]
[v1]
Mon, 9 Feb 2026 11:36:41 UTC (5,785 KB)
[v2]
Thu, 1 Oct 2026 09:13:52 UTC (6,513 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org