arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.20463cs.LG
收藏

漂移模型的长-短流映射视角

A Long-Short Flow-Map Perspective for Drifting Models

School of Interactive Computing(交互计算学院);Georgia Institute of Technology(佐治亚理工学院)

Zhiqi Li, Bo Zhu

AI总结:

本文提出了一种基于长-短流映射分解的漂移模型新视角,通过理论分析和实验验证,实现了对密度演变的对齐,并指出了未来研究的开放问题。

摘要:

本文通过一个半群一致的长-短流映射分解方法重新诠释了漂移模型。我们展示了全局传输过程可以分解为一个长时间跨度的流映射后接一个短时间终端流映射,后者具有闭合形式的最优速度表示。当终端区间长度趋于零时,可以精确恢复漂移场以及为流映射一致性所需的保守冲量项。基于这一视角,我们提出了一种新的似然学习公式,将长-短流映射分解与传输下的密度演变对齐。我们通过理论分析和基准测试的实证评估验证了该框架,并进一步提供了特征空间优化的理论解释,同时指出了几个未来研究中需要解决的开放性问题。

英文摘要:

This paper provides a reinterpretation of the Drifting Model~\cite{deng2026generative} through a semigroup-consistent long-short flow-map factorization. We show that a global transport process can be decomposed into a long-horizon flow map followed by a short-time terminal flow map admitting a closed-form optimal velocity representation, and that taking the terminal interval length to zero recovers exactly the drifting field together with a conservative impulse term required for flow-map consistency. Based on this perspective, we propose a new likelihood learning formulation that aligns the long-short flow-map decomposition with density evolution under transport. We validate the framework through both theoretical analysis and empirical evaluations on benchmark tests, and further provide a theoretical interpretation of the feature-space optimization while highlighting several open problems for future study.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑