# Manifold Attracting Flow Matching: Physics-Constrained Sampling via Continuous Geometric Dynamics

DevFeed: [Manifold Attracting Flow Matching: Physics-Constrained Sampling via Continuous Geometric Dynamics](<https://devfeed.tech/articles/manifold-attracting-flow-matching-physics-constrained-sampling-via-continuous-geometric-dynamics-133139.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16677>)

Author: Guanyu Chen, Jinchen Wang, Shengze Xu, Xingyu Ren, Tieyong Zeng, Antonios Armaou, Dong Ni

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [ode](<https://devfeed.tech/tags/ode.md>), [physics](<https://devfeed.tech/tags/physics.md>), [sampling](<https://devfeed.tech/tags/sampling.md>)

## AI overview

Manifold Attracting Flow Matching (MAFM) enforces previously unseen equality constraints during scientific data generation by evolving a persistent anchor with an ODE that attracts it toward the constraint manifold. Its geometric correction preserves model-guided motion along the manifold while contracting constraint violations in the normal direction, separating model tracking from constraint attraction. The authors report theoretical guarantees for residual contraction and local feedback stabilization, and experiments on geometric and partial differential equation benchmarks show improved generation quality over variant constraint-guided baselines while keeping constraint residuals low. The method is presented as a way to control sampling dynamics without retraining the generative model.

## Source excerpt

arXiv:2611.16677v1 Announce Type: new Abstract: Generating scientific data under previously unseen equality constraints requires satisfying the prescribed constraints without destroying the physical structure learned from data. Existing inference-time approaches often enforce physical constraints by repeatedly correcting predicted endpoints, but such discrete corrections can distort learned generative structure and introduce unstable feedback. In this work, we propose Manifold Attracting Flow Matching (MAFM), which continuously evolves a persistent anchor through an ODE whose dynamics drive it analytically toward the constraint manifold. Specifically, its geometric correction preserves model-guided tangent motion while prescribing the normal motion to contract constraint violations toward the manifold. This construction separates model tracking from constraint attraction and provides theoretical guarantees for residual contraction and local feedback stabilization. Experiments across geometric and partial differential equation benchmarks demonstrate improved generation quality over variant constraint-guided baselines while maintaining low constraint residuals. MAFM provides a flexible framework for coordinating learned physical structure and explicit constraints through controllable sampling dynamics, without retraining the underlying generative model.