---
格式版本: 2
标题: "Composing Flow-Matching Energies with Known Physics: Generation, OOD Detection, and Inversion on PDE Fields"
原文链接: "https://arxiv.org/abs/2608.18004"
发布日期: "2026-08-18"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Tue, 18 Aug 2026 16:53:08 UTC (4,861 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-20T15:53:29+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-20T15:48:16+08:00"
入库时间: "2026-08-20T07:53:29.904Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=model%20fit&searchtype=all"
匹配关键词:
  - "model fit"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 0
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "论文为机器学习流匹配方法研究，与超节点、AI Rack、机柜级AI基础设施无关，无相关技术或商业信号。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T15:59:52+08:00"
AI主题相关性: 0
AI来源权威性: 0
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 0
AI摘要: "研究者提出流匹配模型可通过势诱导速度获得显式能量函数，实现能量修正的PDE场生成、分布外检测和基于能量组合的后验反演采样。该方法无需变分或额外MCMC训练，并在预测-校正采样框架下比流ODE基线降低PDE残差和谱距离。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:21:49.799Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.18004"
---

## Computer Science > Machine Learning

## Title:Composing Flow-Matching Energies with Known Physics: Generation, OOD Detection, and Inversion on PDE Fields

Authors:[Yixuan Sun](https://arxiv.org/search/cs?searchtype=author&query=Sun,+Y), [Anirban Samaddar](https://arxiv.org/search/cs?searchtype=author&query=Samaddar,+A), [Sandeep Madireddy](https://arxiv.org/search/cs?searchtype=author&query=Madireddy,+S)

[View PDF](https://arxiv.org/pdf/2608.18004) [HTML (experimental)](https://arxiv.org/html/2608.18004v1)

> Abstract:Probabilistic modeling of physical fields benefits from both a data-driven prior and known physical structure such as the governing equations. Energy-based models (EBMs) are a natural fit since energies compose additively, which enables augmenting physics information during inference. However, EBMs have been difficult to train and sample from due to the intractable partition function. We show in this work that flow matching models with a potential-induced velocity yield an explicit scalar energy at all transport times, whose gradient is exactly the converted learned score and which recovers the marginal negative log-density at the population optimum. The time-dependent energy functions are obtained purely from the matching regression objective on an independent linear Gaussian interpolation, without a variational form or additional MCMC steps, and the sampling retains the flow ODE. Access to the energy function from a trained model serves three roles: energy-corrected data generation, energy as a scoring function for out-of-distribution (OOD) detection, and energy compositional posterior sampling for inverse problems. In particular, we show the explicit energy permits general MCMC samplers in the predictor-corrector sampling framework, reducing PDE residual and spectral distance compared to the flow ODE baseline. Furthermore, we demonstrate utilizing the data energy and physics-based energy (e.g., PDE residuals) as complementary mechanisms to improve detection accuracy for OOD tasks. In addition, we explore the connection to MCMC-based inference for inverse problems by composing the energy with a quadratic observational likelihood that yields a posterior energy, used as an explicitly chosen family of inference-time targets.

| Subjects: | Machine Learning (cs.LG); Computational Physics (physics.comp-ph) |
| --- | --- |
| Cite as: | [arXiv:2608.18004](https://arxiv.org/abs/2608.18004) \[cs.LG\] |
|  | (or [arXiv:2608.18004v1](https://arxiv.org/abs/2608.18004v1) \[cs.LG\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.18004](https://doi.org/10.48550/arXiv.2608.18004) |

## Submission history

From: Yixuan Sun \[[view email](https://arxiv.org/show-email/1d9be5a8/2608.18004)\]  
**\[v1\]** Tue, 18 Aug 2026 16:53:08 UTC (4,861 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.18004) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))