Generative diffusion and flow models are constructed to remain exactly on the Lorentz-invariant massless N-particle phase space manifold during sampling for particle physics applications.
hub
Flow Matching Guide and Code
32 Pith papers cite this work. Polarity classification is still indexing.
32
Pith papers citing it
abstract
Flow Matching (FM) is a recent framework for generative modeling that has achieved state-of-the-art performance across various domains, including image, video, audio, speech, and biological structures. This guide offers a comprehensive and self-contained review of FM, covering its mathematical foundations, design choices, and extensions. By also providing a PyTorch package featuring relevant examples (e.g., image and text generation), this work aims to serve as a resource for both novice and experienced researchers interested in understanding, applying and further developing FM.
hub tools
representative citing papers
Discrete MeanFlow parameterizes CTMC conditional transition kernels with a boundary-by-construction design to enable exact one-step generation in discrete state spaces.
Asymmetric Langevin Unlearning uses public data to suppress unlearning noise costs by O(1/n_pub²), enabling practical mass unlearning with preserved utility under distribution mismatch.
FlowIQN is a quantile-coupled CFM critic that yields the first explicit Wasserstein-aligned approximate projection for distributional RL, with improved return-distribution accuracy and competitive offline RL performance.
Path-Coupled Bellman Flows use source-consistent Bellman-coupled paths and a lambda-parameterized control-variate to learn return distributions via flow matching, improving fidelity and stability over prior DRL approaches.
OGPP is a particle flow-matching method using orbit-space canonicalization and geometric paths that achieves lower error and fewer steps than prior approaches on 3D benchmarks.
Binomial flows close the gap between continuous flow matching and discrete ordinal data by using binomial distributions to enable unified denoising, sampling, and exact likelihoods in diffusion models.
LeapAlign fine-tunes flow matching models by constructing two consecutive leaps that skip multiple ODE steps with randomized timesteps and consistency weighting, enabling stable updates at any generation step.
TokenLight encodes lighting attributes as tokens in a conditional image generation model trained mostly on synthetic data, enabling precise relighting control and implicit learning of light-scene interactions.
DoMinO reformulates discrete flow matching sampling as an MDP for unbiased RL fine-tuning with new TV regularizers, yielding better enhancer activity and naturalness on DNA design tasks.
DRIFT enables stable offline-to-online fine-tuning of CTMC policies in discrete RL via advantage-weighted discrete flow matching, path-space regularization, and candidate-set approximation.
SF-Flow applies flow matching with a permutation-invariant set encoder and 3D U-Net to reconstruct ATF magnitudes from sparse inputs, showing accurate results up to 1 kHz with faster training than autoencoder baselines.
dFlowGRPO is a new rate-aware RL method for discrete flow models that outperforms prior GRPO approaches on image generation and matches continuous flow models while supporting broad probability paths.
BRICKS creates compositional neural Markov kernels via hybrid transformers and Riemannian Flow Matching on product manifolds to enable zero-shot simulation of radiation-matter interactions across arbitrary material distributions.
Cumulative flow maps unify few-step generative modeling for diffusion and flow models via cumulative transport and parameterization with minimal changes to time embeddings and objectives.
PRTS pretrains VLA models with contrastive goal-conditioned RL to embed goal-reachability probabilities from offline data, yielding SOTA results on robotic benchmarks especially for long-horizon and novel instructions.
Probability Flow Matching learns biophysically consistent stochastic processes for gene regulation from time-resolved single-cell measurements, where only the biophysical versions accurately capture lineage transitions, fate specification, and perturbation responses despite similar data fit.
Occupancy Reward Shaping extracts goal-reaching rewards from world-model occupancy measures using optimal transport, improving offline goal-conditioned RL performance 2.2x on 13 tasks without changing the optimal policy.
Fisher Decorator refines flow policies in offline RL via a local transport map and Fisher-matrix quadratic approximation of the KL constraint, yielding controllable error near the optimum and SOTA benchmark results.
Mixture-of-experts flow matching enables non-autoregressive language models to achieve autoregressive-level quality in three sampling steps, delivering up to 1000x faster inference than diffusion models.
PhyMix unifies a new multi-aspect physics evaluator with implicit policy optimization and explicit test-time correction to produce single-image 3D indoor scenes that are both visually faithful and physically plausible.
MeanFlow uses a derived identity between average and instantaneous velocities to train one-step flow models, achieving FID 3.43 on ImageNet 256x256 with 1-NFE from scratch.
SharpEuler estimates a sharpness profile via finite differences on calibration trajectories, smooths it, and applies a quantile transform to generate adaptive timestep grids that improve Euler sampling quality in flow matching models at fixed budgets.
Numerical benchmarks indicate generative regularizers deliver strong reconstructions in some imaging inverse problem settings but can be unstable or problematic under imperfect conditions compared to variational methods.
citing papers explorer
-
Generative models on phase space
Generative diffusion and flow models are constructed to remain exactly on the Lorentz-invariant massless N-particle phase space manifold during sampling for particle physics applications.
-
Discrete MeanFlow: One-Step Generation via Conditional Transition Kernels
Discrete MeanFlow parameterizes CTMC conditional transition kernels with a boundary-by-construction design to enable exact one-step generation in discrete state spaces.
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
Asymmetric Langevin Unlearning uses public data to suppress unlearning noise costs by O(1/n_pub²), enabling practical mass unlearning with preserved utility under distribution mismatch.
-
Quantile-Coupled Flow Matching for Distributional Reinforcement Learning
FlowIQN is a quantile-coupled CFM critic that yields the first explicit Wasserstein-aligned approximate projection for distributional RL, with improved return-distribution accuracy and competitive offline RL performance.
-
Path-Coupled Bellman Flows for Distributional Reinforcement Learning
Path-Coupled Bellman Flows use source-consistent Bellman-coupled paths and a lambda-parameterized control-variate to learn return distributions via flow matching, improving fidelity and stability over prior DRL approaches.
-
Generative Modeling with Orbit-Space Particle Flow Matching
OGPP is a particle flow-matching method using orbit-space canonicalization and geometric paths that achieves lower error and fewer steps than prior approaches on 3D benchmarks.
-
Binomial flows: Denoising and flow matching for discrete ordinal data
Binomial flows close the gap between continuous flow matching and discrete ordinal data by using binomial distributions to enable unified denoising, sampling, and exact likelihoods in diffusion models.
-
LeapAlign: Post-Training Flow Matching Models at Any Generation Step by Building Two-Step Trajectories
LeapAlign fine-tunes flow matching models by constructing two consecutive leaps that skip multiple ODE steps with randomized timesteps and consistency weighting, enabling stable updates at any generation step.
-
TokenLight: Precise Lighting Control in Images using Attribute Tokens
TokenLight encodes lighting attributes as tokens in a conditional image generation model trained mostly on synthetic data, enabling precise relighting control and implicit learning of light-scene interactions.
-
Discrete Flow Matching Policy Optimization
DoMinO reformulates discrete flow matching sampling as an MDP for unbiased RL fine-tuning with new TV regularizers, yielding better enhancer activity and naturalness on DNA design tasks.
-
Discrete Flow Matching for Offline-to-Online Reinforcement Learning
DRIFT enables stable offline-to-online fine-tuning of CTMC policies in discrete RL via advantage-weighted discrete flow matching, path-space regularization, and candidate-set approximation.
-
SF-Flow: Sound field magnitude estimation via flow matching guided by sparse measurements
SF-Flow applies flow matching with a permutation-invariant set encoder and 3D U-Net to reconstruct ATF magnitudes from sparse inputs, showing accurate results up to 1 kHz with faster training than autoencoder baselines.
-
dFlowGRPO: Rate-Aware Policy Optimization for Discrete Flow Models
dFlowGRPO is a new rate-aware RL method for discrete flow models that outperforms prior GRPO approaches on image generation and matches continuous flow models while supporting broad probability paths.
-
BRICKS: Compositional Neural Markov Kernels for Zero-Shot Radiation-Matter Simulation
BRICKS creates compositional neural Markov kernels via hybrid transformers and Riemannian Flow Matching on product manifolds to enable zero-shot simulation of radiation-matter interactions across arbitrary material distributions.
-
A Few-Step Generative Model on Cumulative Flow Maps
Cumulative flow maps unify few-step generative modeling for diffusion and flow models via cumulative transport and parameterization with minimal changes to time embeddings and objectives.
-
PRTS: A Primitive Reasoning and Tasking System via Contrastive Representations
PRTS pretrains VLA models with contrastive goal-conditioned RL to embed goal-reachability probabilities from offline data, yielding SOTA results on robotic benchmarks especially for long-horizon and novel instructions.
-
Learning biophysical models of gene regulation with probability flow matching
Probability Flow Matching learns biophysically consistent stochastic processes for gene regulation from time-resolved single-cell measurements, where only the biophysical versions accurately capture lineage transitions, fate specification, and perturbation responses despite similar data fit.
-
Occupancy Reward Shaping: Improving Credit Assignment for Offline Goal-Conditioned Reinforcement Learning
Occupancy Reward Shaping extracts goal-reaching rewards from world-model occupancy measures using optimal transport, improving offline goal-conditioned RL performance 2.2x on 13 tasks without changing the optimal policy.
-
Fisher Decorator: Refining Flow Policy via a Local Transport Map
Fisher Decorator refines flow policies in offline RL via a local transport map and Fisher-matrix quadratic approximation of the KL constraint, yielding controllable error near the optimum and SOTA benchmark results.
-
Towards Faster Language Model Inference Using Mixture-of-Experts Flow Matching
Mixture-of-experts flow matching enables non-autoregressive language models to achieve autoregressive-level quality in three sampling steps, delivering up to 1000x faster inference than diffusion models.
-
PhyMix: Towards Physically Consistent Single-Image 3D Indoor Scene Generation with Implicit--Explicit Optimization
PhyMix unifies a new multi-aspect physics evaluator with implicit policy optimization and explicit test-time correction to produce single-image 3D indoor scenes that are both visually faithful and physically plausible.
-
Mean Flows for One-step Generative Modeling
MeanFlow uses a derived identity between average and instantaneous velocities to train one-step flow models, achieving FID 3.43 on ImageNet 256x256 with 1-NFE from scratch.
-
Sharpen Your Flow: Sharpness-Aware Sampling for Flow Matching
SharpEuler estimates a sharpness profile via finite differences on calibration trajectories, smooths it, and applies a quantile transform to generate adaptive timestep grids that improve Euler sampling quality in flow matching models at fixed budgets.
-
A Stability Benchmark of Generative Regularizers for Inverse Problems
Numerical benchmarks indicate generative regularizers deliver strong reconstructions in some imaging inverse problem settings but can be unstable or problematic under imperfect conditions compared to variational methods.
-
Deterministic Decomposition of Stochastic Generative Dynamics
Stochastic generative dynamics admit a transport-osmotic decomposition of the deterministic field, supporting Bridge Matching for interpretable and tunable generation.
-
Exploring Time Conditioning in Diffusion Generative Models from Disjoint Noisy Data Manifolds
Aligning the DDIM forward diffusion process with flow-matching manifold evolution enables high-quality generation without time conditioning, and class-conditional synthesis is possible with an unconditional denoiser by using separate time spaces per class.
-
Efficient Hierarchical Implicit Flow Q-learning for Offline Goal-conditioned Reinforcement Learning
Proposes mean flow policies and LeJEPA loss to overcome Gaussian policy limits and weak subgoal generation in hierarchical offline GCRL, reporting strong results on OGBench state and pixel tasks.
-
Exploring Motion-Language Alignment for Text-driven Motion Generation
MLA-Gen advances text-driven motion synthesis by aligning global motion patterns with fine-grained text semantics and mitigating attention sink effects via new masking techniques.
-
Woosh: A Sound Effects Foundation Model
Woosh is a new publicly released foundation model optimized for high-quality sound effect generation from text or video, showing competitive or better results than open alternatives like Stable Audio Open.
-
A Unified Measure-Theoretic View of Diffusion, Score-Based, and Flow Matching Generative Models
Diffusion, score-based, and flow matching models are unified as instances of learning time-dependent vector fields inducing marginal distributions governed by continuity and Fokker-Planck equations.
-
Generative models for decision-making under distributional shift
Generative models via pushforward maps, Fokker-Planck equations, and Wasserstein geometry enable learning nominal uncertainty, stressed distributions for robustness, and conditional posteriors under distributional shift.
- Mixture Prototype Flow Matching for Open-Set Supervised Anomaly Detection