Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Structured Coupling for Flow Matching

About

Standard flow matching scales well but typically relies on an unstructured source distribution, limiting its ability to learn interpretable latent structure. Latent-variable models, by contrast, capture structure but often sacrifice generative quality. We bridge this gap by proposing Structured Coupling for Flow Matching (SCFM), a cooperative framework that augments flow matching with structured latent representation learning. By introducing structured latent variables and exogenous noise into the source, SCFM jointly learns a structured prior (via latent variable modeling) and a continuous transport map (via flow matching). It uses a shared time-dependent recognition network for both latent variable model variational inference and intermediate-time flow velocity estimation. This yields a structurally informed yet unconditional, simulation-free flow model, where the latent variable model can also assist flow sampling. Empirically, SCFM facilitates unsupervised latent representation learning for clustering, disentanglement and downstream tasks, while remaining competitive with flow matching in sample quality, showing that meaningful structure can be learned without sacrificing generative fidelity.

Xavier Sumba, Carles Balsells-Rodas, Yingzhen Li• 2026

Related benchmarks

TaskDatasetResultRank
ClassificationCIFAR10 (test)
Accuracy66.58
331
ClusteringMNIST (test)
NMI0.8784
136
Image GenerationImageNet 128x128--
74
Disentangled Representation LearningCars3D
FactorVAE0.977
43
DisentanglementShapes3D
FactorVAE Score0.957
26
Image ClassificationImageNet 128
Top-1 Accuracy27.96
4
Image GenerationCIFAR-10
FID (50k)2.117
2
Showing 7 of 7 rows

Other info

Follow for update