Flow Matching

Ask about this topic

Overview

Flow matching learns a velocity field that moves samples from a simple source, often Gaussian noise, toward a target data distribution. Conditional flow matching trains the field on paths between source and data samples, while optimal-transport couplings and rectified paths aim to make those trajectories easier to learn and integrate with an ordinary differential equation solver. Extensions handle discrete data, curved manifolds and symmetries, physical constraints, and applications from image and speech generation to robotics, scientific forecasting, and inverse problems. A central challenge is getting high fidelity and diverse, reliable samples with few solver steps; theory on convergence and stability is advancing, but the best paths, source distributions, and schedules depend on the task.

Latest papers

CardsList

663 papers123 papers in the last four weeks, up 68% on the four weeks before. 0.7% of all new papers.

  1. Coupled Guidance for Flow Matching

    Oct 29, 2026Mateo Clémente, Léo Maxime Brunswic, Rui Heng Yang +1Flow MatchingFlow-Based Generative Modeling

    Huawei

  2. Beyond Reachability: When Edit Order Preserves Couplings in Graph Flow Matching

    Oct 28, 2026Zilin Li, Zixing SongGraph GenerationFlow Matching

    Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University

  3. NAGO: Noise-Aware Generative Operator via Flow Matching

    Oct 28, 2026Guanyu Chen, Shengze Xu, Pengwei Liu +5Flow-Based Generative ModelingFlow Matching

    Zhejiang University · The Chinese University of Hong Kong · The Hong Kong Polytechnic University +1

  4. Physics-Guided Flow Matching: A Generative AI Framework for Global Total Electron Content Forecasting

    Oct 27, 2026Si-Tong Yang, Feng Ye, Xian Lu +2Spatiotemporal ForecastingPhysics-Informed Generative Modeling

    Department of Industrial Engineering, Clemson University, Clemson, SC, USA · Department of Physics and Astronomy, Clemson University, Clemson, SC, USA · Department of Physics, Utah State University, Logan, UT, USA

  5. HELICS: Biobank-scale Conditional Synthetic Genome Generation via Latent Flow Matching

    Oct 26, 2026Ahmad Abdel-Azim, Xihong LinSynthetic Data GenerationGenomics

    Department of Biostatistics, Harvard University, Boston, MA 02135 · Department of Statistics, Biostatistics, Harvard University, Boston, MA 02135

  6. Beyond the Estimate: Wiener-Informed Decoding for Easier Training of Pixel-Space Flow Models

    Oct 26, 2026Ziqiao Wang, Yang YouFlow-Based Generative ModelingImage Generation

    NUS HPC-AI Lab

  7. Nowcasting aerosol particle size distributions from PM2.5 with physics-regularized flow matching

    Oct 24, 2026Zhi-Song Liu, Jenni Köykkä, Wenqing Peng +3Physics-Informed Generative ModelingAir Quality Prediction

    Department of Computational Engineering, LUT University, Finland · Atmospheric Modelling Center, Lahti (AMC-Lahti), Finland · Institute for Atmospheric and Earth System Research (INAR), University of Helsinki, Finland

  8. Climate downscaling with likelihood functional flow matching

    Oct 24, 2026Zhi-Song Liu, Helmi Toropainen, Ammar Kheder +1Climate DownscalingFlow Matching

    Department of Computational Engineering, LUT University, Lappeenranta, Finland. · Institute for Atmospheric and Earth System Research (INAR), University of Helsinki, Helsinki, Finland. · Atmospheric Modelling Center, Lahti (AMC-Lahti), Lahti, Finland.

  9. Flow Mesh: A Velocity-Regularized Flow Matching Framework for Stable Distributional Modeling of Physical Fields on Unstructured Meshes

    Oct 23, 2026Siddharth Verma, Jaewan Park, Jay Phil Yoo +3Generative ModelingFlow-Based Generative Modeling

    Department of Mechanical Engineering, Indian Institute of Technology Bombay, Mumbai, India · National Center for Supercomputing Applications, Urbana, IL, USA · Grainger College of Engineering, Mechanical Science & Engineering Department, University of Illinois Urbana-Champaign, Urbana, IL, USA +3

  10. Cross-Scenario Controller Parameter Transformation from Real-World Driving Data through Flow Matching

    Oct 23, 2026Christian Zink, Maik Pfefferkorn, Eric Lenz +2Autonomous DrivingFlow Matching

    Control and Cyber-Physical Systems Laboratory, Technical University of Darmstadt, Darmstadt, Germany · Driver Assistance Systems Predevelopment, Volkswagen AG, Wolfsburg, Germany

  11. Representation Learning via Composite Subspace Flow Matching

    Oct 23, 2026José Afonso, Mário FigueiredoSelf-Supervised Visual Representation LearningFlow Matching

    Instituto Superior Técnico, Lisboa, Portugal · Instituto de Telecomunicações, Instituto Superior Técnico, Lisboa, Portugal

  12. Compatible Likelihoods for Flow Matching on Manifolds

    Oct 22, 2026Lucas Ng, Georgios Batzolis, Mark GirolamiFlow MatchingRiemannian Flow Matching

    University of Cambridge · Devotion AI Labs Ltd.

  13. VisFlow: Interpretable Flow Matching for Visibility Reconstruction, Characterization, and Recognition

    Oct 22, 2026Rahul Moorthy, Volkan IslerFlow MatchingGraph Reconstruction

  14. GLOW-FM: Efficient Generative Low-Light Image Enhancement with One-Step Latent Flow Matching

    Oct 22, 2026Ariel Lapid, Yuval Haitman, Reuven Peretz +2Low-Light Image EnhancementOne-Step Generative Modeling

    Arm AI Research (AAIR), Arm Ltd

  15. When One Step Is Enough: An Empirical Study of Euler Integration in Flow-Matching Policies

    Oct 22, 2026Shivakanth Sujit, Luca Nunziante, Rousslan Fernand Julien Dossa +1Flow MatchingEfficient VLA Model Inference

    Araya Inc., Tokyo

  16. Learning Domain Trajectories with Flow Matching for Gradual Domain Adaptation

    Oct 22, 2026Yubo Huang, Yushe Cao, Zixi Wang +4Domain AdaptationUnsupervised Domain Adaptation

    Key Laboratory of Transportation Tunnel Engineering, Ministry of Education, Southwest Jiaotong University, Chengdu, China · Tsinghua University, Beijing, China · School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, China +2

  17. Velocity Ambiguity Profiles: Time-Resolved Bayes-Risk Diagnostics for Flow Matching

    Oct 21, 2026Yunpeng Mei, Xiaowen Zhu, Chenyu Wang +5Flow Matching

    Beijing Institute of Technology · ByteDance

  18. Compositional neural networks for multimodal X-ray and neutron scattering inversion

    Oct 21, 2026Valentin Munteanu, Alexander Hinderhofer, Dmitry Lapkin +2Inverse ProblemsUncertainty-Aware Multimodal Fusion

    Institute of Applied Physics, University of Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, Germany · Cluster of Excellence ”Machine Learning” - new perspectives for science, University of Tübingen, Germany

  19. StyleStream 2.0: Fast and Controllable Streaming Voice Style Conversion

    Oct 20, 2026Yisi Liu, Nicholas Lee, Gopala K. AnumanchipalliVoice ConversionZero-Shot Voice Conversion

    University of California, Berkeley

  20. Minimax-Optimal Predictive Flow Matching under Distribution Drift

    Oct 20, 2026Aadithya Srikanth, Vaneet AggarwalFlow MatchingDynamic Regret Minimization

    Purdue University, West Lafayette

  21. From Short Clips to Real-Time Streams: Identity-Aware Supervision and Context-Efficient Generation for Video Face Swapping

    Oct 19, 2026Zekai Luo, Yuwei Liang, Ruiqi Ma +8Real-Time Interactive Video GenerationIdentity-Preserving Video Generation

    Zhejiang University · Zhejiang University of Technology

  22. HP-PnP-Flow: Hybrid Time Scheduling and Projection for Plug-and-Play Image Restoration via Flow Matching

    Oct 19, 2026Yong Li, Xiaojie Wang, Xiaoyan ZhangImage RestorationImage Denoising

    School of Mathematics and Statistics, HNP-LAMA, Central South University, Changsha 410083, Hunan, China

  23. DeOverlay: A Benchmark and Source-Anchored Flow for Video Overlay Removal

    Oct 19, 2026Deshui Miao, Yuhang Ma, Shaoyong Jia +4Flow Matching

    ByteDance · Southern University of Science and Technology

  24. PnP-CF: A Consensus Flow-Matching Solver for General Inverse Problems

    Oct 19, 2026Dibyanshu Kumar, Magda GregorovaInverse ProblemsFlow Matching

    TH Würzburg-Schweinfurt, CAIRO, University of Würzburg

  25. HCFM: Helmholtz Component Flow Matching for Time Series Anomaly Detection

    Oct 18, 2026Shyam Sundar Saravanan, Tie Luo, Sanjay MadriaTime-Series Anomaly DetectionUnsupervised Anomaly Detection

    Department of Computer Science, Missouri University of Science and Technology, Rolla, MO, USA · Department of ECE, University of Kentucky, Lexington, KY, USA

  26. CorrFlow: Self-Correcting Flow Matching

    Oct 17, 2026Hangwei Zhang, Nathan Rouillé, Chuanxia ZhengFlow MatchingFlow-Based Generative Modeling

    PVG, Nanyang Technological University · Télécom Paris, Institut Polytechnique de Paris

  27. BEM: Learning Bregman Divergences for Concept Erasure in Flow Matching Models

    Oct 17, 2026Mohamed-Hicham Leghettas, Markus PüschelDiffusion Model UnlearningFlow Matching

    Department of Computer Science, ETH Zurich, Switzerland

  28. GliomaINK: A Tumor and Anatomy Conditioned Flow Matching Framework for 3D MRI Sequence Synthesis in Glioma Patients

    Oct 16, 2026Moinak Bhattacharya, Annie Singh, Mohammad Chaudhry +2Medical Image SynthesisBrain MRI

    Department of Radiology, Columbia University Irving Medical Center, NY, US · Atal Bihari Vajpayee Institute of Medical Sciences, New Delhi, India · Department of Biomedical Informatics, Stony Brook University, NY, US

  29. Rethinking Classifier-Free Guidance for TI2I Flow Matching

    Oct 15, 2026Andreas Floros, Elena Koritska, Teofilo de Campos +4Classifier-Free GuidanceText-Guided Image Editing

    Imperial College London · Microsoft

  30. ControlFlow3D: Distilling Multi-View Knowledge into Latent Flow Matching for Point Cloud Upsampling

    Oct 15, 2026Yuang Liu, Zhi Zuo, Zhengkai Zhao +4Point Cloud LearningFlow Matching

    Nanjing University of Aeronautics and Astronautics · Weidian Corporation · Lingnan University