Sinkhorn-Drifting Generative Models

• URL: http://arxiv.org/abs/2603.12366v1
• Date: Thu, 12 Mar 2026 18:33:36 GMT
• Title: Sinkhorn-Drifting Generative Models
• Abstract summary: "Drifting" generative dynamics and gradient flows induced by Sinkhorn divergence.<n>Drifting reduces sensitivity to kernel temperature and improves one-step generative quality.<n>These theoretical gains translate to strong low-temperature improvements in practice.
• Score: 27.763380775285466
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We establish a theoretical link between the recently proposed "drifting" generative dynamics and gradient flows induced by the Sinkhorn divergence. In a particle discretization, the drift field admits a cross-minus-self decomposition: an attractive term toward the target distribution and a repulsive/self-correction term toward the current model, both expressed via one-sided normalized Gibbs kernels. We show that Sinkhorn divergence yields an analogous cross-minus-self structure, but with each term defined by entropic optimal-transport couplings obtained through two-sided Sinkhorn scaling (i.e., enforcing both marginals). This provides a precise sense in which drifting acts as a surrogate for a Sinkhorn-divergence gradient flow, interpolating between one-sided normalization and full two-sided Sinkhorn scaling. Crucially, this connection resolves an identifiability gap in prior drifting formulations: leveraging the definiteness of the Sinkhorn divergence, we show that zero drift (equilibrium of the dynamics) implies that the model and target measures match. Experiments show that Sinkhorn drifting reduces sensitivity to kernel temperature and improves one-step generative quality, trading off additional training time for a more stable optimization, without altering the inference procedure used by drift methods. These theoretical gains translate to strong low-temperature improvements in practice: on FFHQ-ALAE at the lowest temperature setting we evaluate, Sinkhorn drifting reduces mean FID from 187.7 to 37.1 and mean latent EMD from 453.3 to 144.4, while on MNIST it preserves full class coverage across the temperature sweep. Project page: https://mint-vu.github.io/SinkhornDrifting/

Related papers

• Engineering exact mobility edges in quasiperiodic Aharonov-Bohm chains [0.09558392439655013]
  We investigate localization phenomena and exact mobility edges in a quasiperiodic Aharonov-Bohm chain.<n>In the diagonal limit, the coupling to the auxiliary sublattice generates hyperbolic mobility edges that exhibit a sign-changing divergence.<n>In the off-diagonal regime, the interference between quasiperiodic hopping and indirect tunneling through the auxiliary sublattice gives rise to exact anomalous mobility edges.
  arXiv  Detail & Related papers  (2026-08-23T07:54:01Z)
• Finite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Models [5.256913964154542]
  We propose and analyze a conservative drifting method for one-step generative modeling.<n>The method replaces the original displacement-based drifting velocity by a kernel density estimator (KDE)-gradient velocity.<n>We also analyze the non-conservative drifting method with Laplace kernel, corresponding to the original displacement-based velocity proposed in Deng et al., 2026.
  arXiv  Detail & Related papers  (2026-05-21T17:49:09Z)
• Attraction, Repulsion, and Friction: Introducing DMF, a Friction-Augmented Drifting Model [37.847379736905175]
  We show that a linearly-scheduled friction gives a finite-horizon bound on the error trajectory.<n>Our friction-augmented model, DMF, matches or exceeds Optimal Flow Matching on FFHQ coefficient adult-to-child domain translation.
  arXiv  Detail & Related papers  (2026-04-20T12:47:27Z)
• OmniForcing: Unleashing Real-time Joint Audio-Visual Generation [51.031726911410594]
  Recent joint audio-visual diffusion models achieve remarkable generation quality but suffer from high latency.<n>We propose OmniForcing, the first framework to distill an offline, dual-stream bidirectional diffusion model into a high-fidelity streaming autore generator.
  arXiv  Detail & Related papers  (2026-03-12T08:17:36Z)
• Generative Drifting is Secretly Score Matching: a Spectral and Variational Perspective [33.2175112734242]
  Generative Modeling via Drifting has recently achieved state-of-the-art one-step image generation through a kernel-based drift operator.<n>We make the following observation: emphunder a Gaussian kernel, the drift operator is exactly a score difference on smoothed distributions.
  arXiv  Detail & Related papers  (2026-03-10T17:30:35Z)
• A Unified View of Drifting and Score-Based Models [92.51878018325031]
  We make our relationship to the score-matching principle behind diffusion models precise by showing that drifting admits a score-based formulation on kernel-smoothed distributions.<n>We derive an exact decomposition for general radial kernels, and for the Laplace kernel we prove rigorous error bounds showing that drifting remains an accurate proxy for score matching in low-temperature and high-dimensional regimes.
  arXiv  Detail & Related papers  (2026-03-08T07:41:36Z)
• Symmetry-protected topology and deconfined solitons in a multi-link $\mathbb{Z}_2$ gauge theory [45.88028371034407]
  We study a $mathbbZ$ lattice gauge theory defined on a multi-graph with links that can be visualized as great circles of a spherical shell.<n>We show that this leads to state-dependent tunneling amplitudes underlying a phenomenon analogous to the Peierls instability.<n>By performining a detailed analysis based on matrix product states, we prove that charge deconfinement emerges as a consequence of charge-fractionalization.
  arXiv  Detail & Related papers  (2026-03-02T22:59:25Z)
• Contraction and entropy production in continuous-time Sinkhorn dynamics [0.6423239719448169]
  We give an exact identity for the entropy production rate of the Sinkhorn flow, which was previously known only to be nonpositive.<n>We show that the flow induces a reversible Markov dynamics on the target marginal as an Onsager gradient flow.<n>We give for illustration two immediate practical use-cases for the Sinkhorn LSI: as a design principle for the latent space in which generative models are trained, and as a stopping algorithm for discrete-time algorithms.
  arXiv  Detail & Related papers  (2025-10-14T15:32:15Z)
• On the Wasserstein Convergence and Straightness of Rectified Flow [54.580605276017096]
  Rectified Flow (RF) is a generative model that aims to learn straight flow trajectories from noise to data.<n>We provide a theoretical analysis of the Wasserstein distance between the sampling distribution of RF and the target distribution.<n>We present general conditions guaranteeing uniqueness and straightness of 1-RF, which is in line with previous empirical findings.
  arXiv  Detail & Related papers  (2024-10-19T02:36:11Z)
• Wasserstein Mirror Gradient Flow as the limit of the Sinkhorn Algorithm [2.7240657895633436]
  We prove that the sequence of marginals obtained from the iterations of the Sinkhorn algorithm converges to an absolutely continuous curve on the $2$-Wasserstein space.<n>This limit, which we call the Sinkhorn flow, is an example of a Wasserstein mirror gradient flow.
  arXiv  Detail & Related papers  (2023-07-31T06:11:47Z)
• Distributional Reinforcement Learning with Regularized Wasserstein Loss [16.042828467744197]
  We propose textitSinkhorn distributional RL (SinkhornDRL), which leverages Sinkhorn divergence, a regularized Wasserstein loss, to minimize the difference between current and target Bellman return distributions. Empirically, we show that SinkhornDRL consistently outperforms or matches existing algorithms on the Atari games suite and particularly stands out in the multi-dimensional reward setting.
  arXiv  Detail & Related papers  (2022-02-01T21:27:51Z)
• Accelerating Nonconvex Learning via Replica Exchange Langevin Diffusion [67.66101533752605]
  Langevin diffusion is a powerful method for non- optimization. We propose replica exchange, which swaps Langevin diffusions with different temperatures. By discretizing the replica exchange Langevin diffusion, we obtain a discretetime algorithm.
  arXiv  Detail & Related papers  (2020-07-04T02:52:11Z)
• Turbulent scalar flux in inclined jets in crossflow: counter gradient transport and deep learning modelling [0.0]
  Investigation of turbulent scalar mixing sheds light onto previously observed but unexplained phenomenon of negative turbulent diffusivity. Deep learning approach for modelling turbulent scalar flux by adapting the tensor basis neural network previously developed to model Reynolds stresses. Results show significant improvement compared to a simple model, particularly where cross-gradient effects play an important role in turbulent mixing.
  arXiv  Detail & Related papers  (2020-01-14T03:02:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.