AdjointFlow：用物理驱动的传输加速科学反问题求解

技术上，AdjointFlow 基于 stochastic interpolants，把已知物理模型直接注入生成传输过程：用基于伴随（adjoint）的数据相关初始化缩短到目标分布的传输距离，修改传输速度使其在目标分布高度集中时保持正则，并将初始化直接内建到网络参数化的速度场中。可选的推理时物理模型查询细化机制无需重训练即可进一步提升数据一致性。

该框架针对的是经典求解器在非凸问题和昂贵物理模型反复求值中的瓶颈，也弥补了纯数据驱动条件模型忽略物理、传输映射病态的缺陷。对地震成像、无损检测、天体物理等领域的科研计算团队，这提供了一条在精度与计算成本之间更平衡的路线。

AdjointFlow: Accelerating Scientific Inverse Problems with Physics-Based Transport

Inverse problems are ubiquitous across scientific applications ranging from seismic imaging to astrophysics to nondestructive testing, yet classical solvers often struggle with nonconvexity and repeated evaluations of expensive physical models. Diffusion models can amortize this computation offline, but largely data-driven conditional models underutilize known physics and suffer from ill-conditioned transport maps when the target distribution is concentrated. Plug-and-play (PnP) methods explicitly leverage the physical model to guide a learned prior, improving data consistency but reintroducing costly iterative computations. We instead propose AdjointFlow, a framework built on stochastic interpolants that injects the known physics into the generative transport itself. We design a data-dependent, adjoint-based initialization to reduce transport distance to the target, modify the transport velocity to remain regular as the endpoint distributions concentrate, and parameterize the network to build the initialization directly into the learned velocity. Together, these enable accurate reconstruction at substantially lower cost than generative models starting from Gaussian noise. Experiments on inverse scattering, seismic imaging, and black hole imaging show that AdjointFlow consistently produces more accurate reconstructions than conditional diffusion models while being up to five orders of magnitude faster at inference than PnP methods. Optional inference-time refinement that queries the physical model can further improve data consistency, matching or exceeding the accuracy of PnP methods without retraining.

本条情报归属 AITOP「论文」核心栏目，重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：AdjointFlow、扩散模型、科学计算、随机插值，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。

信源核验说明：本文原始线索采编自「arXiv cs.LG」（@Ethan Hanold, Nicholas M. Boffi, Qin Li, Leonardo Zepeda-Núñez），经 AITOP 多维评分管线结构化提炼。