Point interpolation and distributional universality in flow-based generative models
Authors: Antonio Álvarez-López, Jingpu Cheng, Liyan Zhang, Qianxiao Li
Organizations: Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798. · Department of Mathematics, National University of Singapore, 10 Lower Kent Ridge Road, Singapore 119076. · Institute for Functional Intelligent Materials, National University of Singapore, 10 Lower Kent Ridge Road, Singapore 119076.
Motivated by the expressive power of neural ordinary differential equations in flow-based generative modeling, we study the approximation of probability measures by pushforwards generated by parameterized families of Lipschitz vector fields. For Wasserstein distances, we establish a quantitative relationship between finite-sample point interpolation and distributional approximation, and derive universality criteria based on Lp-approximation of transport maps. In particular, affine-invariant families containing a non-affine vector field are universal from compactly supported absolutely continuous base measures. For total variation, we show that coordinate-wise dynamics are insufficient in dimensions d≥2 and establish sufficient conditions based on one-dimensional Sobolev approximation and shear-type mixing. We further examine the role of point interpolation, establishing when different regimes relate to one another and when approximate interpolation can be made exact.
Figures & tables
Figure 1: Logical implications among the main properties we study. Green , pink , and blue colors indicate TV approximation, point interpolation, and Wp approximation, respectively, while highlighted boxes mark the properties featured in our main results. Solid arrows denote implications, double-headed arrows indicate equivalence, and red arrows with a cross indicate strict non-implications. Finally, dashed arrows represent conditional implications that hold under the assumptions stated in the referenced results.
Figure 3: Illustration of the four-step construction in the proof of Theorem 4.7 for d=2 .
Flow-based generative modeling learns probability transports from simple noise to data for fast sampling, density estimation, and controllable generation.
We formulate flow matching on probability measures over probability measures. Directly from the Wasserstein-on-Wasserstein (WoW) continuity equation, we show that admissible nested outer and inner couplings induce a valid metameasure flow and non-local velocity field. Validity requires neither optimal plans nor a smooth manifold. Practically, this opens up the design space of multisample flow matching: exact Wasserstein plans are distinguished by energy and straightness, while sliced and lazy linear plans yield scalable alternatives. Point-cloud experiments, including controlled Bures-Wasserstein and Euclidean minibatch baselines, demonstrate the resulting geometry-cost trade-off.
Moritz Piening, Richard Duong, Gabriele Steidl
Institut für Mathematik, Technische Universität Berlin, Germany
We survey continuous-time generative modeling methods based on transporting a simple reference distribution to a data distribution via stochastic or deterministic dynamics. We present a unified framework in which diffusion models, score-based generative models, and flow matching are instances of learning a time-dependent vector field that induces a family of marginals (ρt)t∈[0,1] governed by continuity and Fokker-Planck equations. Such a unified theory is timely because these methods are converging methodologically, yet fragmented notation and competing derivations continue to obscure their shared structure and the practical tradeoffs governing sampling, stability, and computation. Within this framework, we (i) derive reverse-time sampling for diffusion and score-based models as controlled stochastic dynamics, (ii) show that the probability flow ODE yields identical marginals and connects diffusion to likelihood-based normalizing flows, and (iii) interpret flow matching as direct regression of the velocity field under a chosen interpolation, clarifying when it coincides with or differs from score-based training. We compare objectives, sampling schemes, and discretization errors under unified notation, discuss connections to Schrodinger bridges and entropic optimal transport, and summarize theoretical guarantees and open problems on approximation, stability, and scalability.
Aditya Ranganath, Mukesh Singhal
Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore, CA 94551, USA · Department of Electrical Engineering University of California Merced, CA 95343, USA
Flow Matching is a powerful framework for learning transport maps between probability distributions. Yet its standard single-parameter formulation is not designed to capture multi-parameter variations where the resulting transport should be path-independent. Path independence is crucial because it ensures that transformations depend only on the initial and target distributions, not on the specific path. In this work, we introduce Path-independent Flow Matching (PiFM), a method for learning vector fields whose induced flows yield path-independent transport between distributions. We show that PiFM generalizes Flow Matching to higher-dimensional parameter domains while enforcing structural conditions that ensure consistency of composed transformations. In addition, we show that, under suitable assumptions, PiFM approximates the Wasserstein barycenter, linking the framework to a notion of distributional interpolation. To enable practical training, we propose a tractable, simulation-free objective that regresses onto multi-parameter conditional probability paths. We showcase empirically that PiFM outperforms other approaches on both synthetic and real world data in interpolating path-independent trajectories and generating desired out of distribution samples.
Francisco Téllez, AmirHossein Zamani, Philippe Martin +5
Université de Montréal, Mila · Concordia University, Mila · Royal Military College of Canada +1