Flow-Based Generative Modeling

Follow this topic Ask about this topic

Flow-based generative modeling represents generation as a transformation of probability mass, often by integrating a continuous-time velocity field from noise to data or composing invertible normalizing-flow layers. Flow matching and rectified flow train velocity fields by regression along selected paths; optimal-transport couplings, stochastic interpolants, discrete Markov flows, and flow maps extend the toolkit to different data types and sampling budgets. Researchers apply these methods to images, language, molecules, physical fields, and robot actions, assessing fidelity and diversity alongside likelihood, constraint satisfaction, calibration, and runtime. A central challenge is achieving one- or few-step generation without sacrificing quality or mode coverage, while improving path design, robustness, geometric and physical consistency, and theoretical guarantees.

Latest papers

563 papers125 papers in the last four weeks, up 131% on the four weeks before. 0.7% of all new papers.

Deterministic Decomposition of Stochastic Generative Dynamics
May 9, 2026Xingyu Song, Yuan Mei, Naoya TakeishiFlow Matching Stochastic Differential Equations
1The University of Tokyo · 2Zhejiang University · *Visiting student at The University of Tokyo from Zhejiang University.
Generalized Wasserstein Flow Matching: Nested Couplings for Point Cloud Generation
May 8, 2026Moritz Piening, Richard Duong, Gabriele SteidlFlow Matching Generative Modeling
Institut für Mathematik, Technische Universität Berlin, Germany
Exploring and Exploiting Stability in Latent Flow Matching
May 8, 2026Rania Briq, Michael Kamp, Ohad Fried +2Flow Matching Efficient Inference
Forschungszentrum Jülich · Technical University Dortmund · Lamarr Institute +4
What Time Is It? How Data Geometry Makes Time Conditioning Optional for Flow Matching
May 8, 2026Alec Helbling, Sebastian Gutierrez Hernandez, Benjamin Hoover +2Flow Matching Flow-Based Generative Modeling
Georgia Tech · IBM Research
Normalizing Trajectory Models
May 8, 2026Jiatao Gu, Tianrong Chen, Ying Shen +3Normalizing Flows Diffusion Model Distillation
Apple · UIUC
Flow-OPD: On-Policy Distillation for Flow Matching Models
May 8, 2026Zhen Fang, Wenxuan Huang, Yu Zeng +8Diffusion Model Alignment Flow Matching
University of Science and Technology of China · University of California, Los Angeles · 3The Chinese University of Hong Kong +1
STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation
May 8, 2026Ying Shen, Tianrong Chen, Yuan Gao +6Unified Multimodal Models Multimodal Generation
UIUC · Apple
Trajectory as the Teacher: Few-Step Discrete Flow Matching via Energy-Navigated Distillation
May 8, 2026Amin Karimi Monsefi, Dominic Culver, Nikhil Bhendawade +3Flow Matching Diffusion Language Model Inference
1The Ohio State University · 2Apple
Consistency Regularised Gradient Flows for Inverse Problems
May 8, 2026Alessio Spagnoletti, Tim Y. J. Wang, Marcelo Pereyra +1Latent Diffusion Models Wasserstein Gradient Flows
Laboratoire MAP5, UMR 8145, Université Paris Cité, CNRS · Department of Mathematics, Imperial College London · Heriot-Watt University, MACS & Maxwell Institute
Scaling Categorical Flow Maps
May 8, 2026Oscar Davis, Anastasiia Filippova, Pierre Ablin +4Language Model Scaling Flow Map Learning
Apple · University of Oxford
Flow Matching for Count Data
May 8, 2026Ganchao Wei, John PearsonFlow Matching Synthetic Data Generation
Department of Neurobiology Department of Statistical Science Duke University Durham, NC, USA · Department of Neurobiology Department of Electrical and Computer Engineering Duke University Durham, NC, USA
Structured Coupling for Flow Matching
May 8, 2026Xavier Sumba, Carles Balsells-Rodas, Yingzhen LiFlow Matching Disentangled Representation Learning
Imperial College London
Flow-Direct: Feedback-Efficient and Reusable Guidance for Flow Models via Non-Parametric Guidance Field
May 8, 2026Kim Yong Tan, Yueming Lyu, Ivor Tsang +1Diffusion Model Guidance Black-Box Optimization
College of Computing and Data Science, Nanyang Technological University, Singapore · Centre for Frontier AI Research, Agency for Science, Technology and Research, Singapore · Institute of High Performance Computing, Agency for Science, Technology and Research, Singapore
Conservative Flows: A New Paradigm of Generative Models
May 7, 2026Eshed Gal, Md Shahriar Rahim Siddiqui, Moshe Eliasof +1Langevin Dynamics Generative Flow Networks
The University of British Columbia Vancouver, Canada · University of Cambridge Cambridge, United Kingdom
A Unified Measure-Theoretic View of Diffusion, Score-Based, and Flow Matching Generative Models
May 7, 2026Aditya Ranganath, Mukesh SinghalFlow Matching Generative Modeling
Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore, CA 94551, USA · Department of Electrical Engineering University of California Merced, CA 95343, USA
Improved techniques for fine-tuning flow models via adjoint matching: a deterministic control pipeline
May 7, 2026Zhengyi Guo, Jiayuan Sheng, David D. Yao +1Fine-Tuning Flow-Based Generative Modeling
FREPix: Frequency-Heterogeneous Flow Matching for Pixel-Space Image Generation
May 7, 2026Mingfeng Lin, Jiakun Chen, Liang Han +1Flow Matching Conditional Image Generation
Harbin Institute of Technology (Shenzhen)
Flow Matching with Arbitrary Auxiliary Paths
May 7, 2026Xin Peng, Ang GaoConditional Flow Matching Flow-Based Generative Modeling
School of Physical Science and Technology, Beijing University of Posts and Telecommunications, Beijing, China
A Flow Matching Algorithm for Many-Shot Adaptation to Unseen Distributions
May 7, 2026Tyler Ingebrand, Ruihan Zhao, Kushagra Gupta +3Flow Matching Flow-Based Generative Modeling
University of Texas at Austin
P-Guide: Parameter-Efficient Prior Steering for Single-Pass CFG Inference
May 7, 2026Xin Peng, Ang GaoFlow Matching Classifier-Free Guidance
School of Physical Science and Technology, Beijing University of Posts and Telecommunications, Beijing, China
Physical Fidelity Reconstruction via Improved Consistency-Distilled Flow Matching for Dynamical Systems
May 7, 2026Sicheng Ma, Tianyue Yang, Xiuzhe Wu +1Flow Matching One-Step Generative Modeling
Department of Physics University of Cambridge · The Center for Computational Science University College London · Department of Computer Science University College London
SDFlow: Similarity-Driven Flow Matching for Time Series Generation
May 7, 2026Wei Li, Shibo Feng, Pengcheng Wu +3Flow Matching Time Series Generation
Shanghai Jiao Tong University · Shanghai University · Nanyang Technological University +3
Spherical Flows for Sampling Categorical Data
May 7, 2026Jannis Chemseddine, Gregor Kornhardt, Gabriele SteidlFlow Matching Synthetic Data Generation
Technische Universität Berlin
HDFlow: Hierarchical Diffusion-Flow Planning for Long-horizon Tasks
May 6, 2026Nandiraju Gireesh, Yuanliang Ju, Chaoyi Xu +3Hierarchical Planning Long-Horizon Robotic Manipulation
1Peking University · University of Toronto. · 2Galbot
A Few-Step Generative Model on Cumulative Flow Maps
May 5, 2026Zhiqi Li, Duowen Chen, Yuchen Sun +1One-Step Generative Modeling Flow-Based Generative Modeling
Georgia Institute of Technology, Atlanta, United States of America
PerFlow: Physics-Embedded Rectified Flow for Efficient Reconstruction and Uncertainty Quantification of Spatiotemporal Dynamics
May 5, 2026Hao Zhou, Rui Zhang, Han Wan +1Physics-Informed Generative Modeling Uncertainty Quantification
Gaoling School of Artificial Intelligence, Renmin University of China
Generative Modeling with Orbit-Space Particle Flow Matching
May 4, 2026Sinan Wang, Jinjin He, Shenyifan Lu +3Flow Matching One-Step Generative Modeling
Georgia Institute of Technology, USA
Stable GFlowNets with Probabilistic Guarantees
May 3, 2026Zengxiang Lei, Ananth Shreekumar, Jonathan Rosenthal +6Generative Flow Networks Flow-Based Generative Modeling
Lyles School of Civil and Construction Engineering, Purdue University, IN, USA · Department of Computer Science, Purdue University, IN, USA · Computer Science and Engineering, University of California, Santa Cruz, USA
CoFlow: Coordinated Few-Step Flow for Offline Multi-Agent Decision Making
May 2, 2026Guowei Zou, Haitao Wang, Beiwen Zhang +2Flow Matching Multi-Agent Coordination
Sun Yat-sen University
Posterior Augmented Flow Matching
May 1, 2026George Stoica, Sayak Paul, Matthew Wallingford +6Flow Matching Conditional Image Generation
Georgia Tech · University of Washington · Hugging Face +2

May 9, 2026J/K move · Enter open · S save

Deterministic Decomposition of Stochastic Generative Dynamics

Xingyu Song, Yuan Mei, Naoya Takeishi

1The University of Tokyo · 2Zhejiang University · *Visiting student at The University of Tokyo from Zhejiang University.

Flow Matching Stochastic Differential Equations Generative Modeling Flow-Based Generative Modeling

Open paper AskPDF arXiv

Modern generative models can be understood as probability transport from a simple base distribution to a target data distribution. Deterministic transport models offer tractable velocity-field parameterizations, whereas stochastic generative models capture richer density evolution through drift and diffusion. Yet when stochastic dynamics are described through deterministic velocity fields, the effects of drift and diffusion are often compressed into a single effective field, obscuring the distinct roles of deterministic evolution and stochastic fluctuation. In this work, we show that the deterministic field

b_t

of a stochastic generative process admits a natural transport--osmotic decomposition that separates deterministic transport from stochastic, diffusion-induced effects:

b_t = u_t + d_t

, where

u_t

governs marginal probability transport and

d_t

captures an osmotic effect induced by diffusion and determined by the marginal score. Based on this decomposition, we propose Bridge Matching, a flow-based framework for learning decomposed generative dynamics through both marginal and conditional formulations. In generative modeling experiments, we recombine the learned components as

b_t = u_t + λ_d d_t

, showing that the proposed decomposition enables interpretable and controllable sampling by adjusting the osmotic contribution in probability transport.