{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T20:58:55Z","timestamp":1776977935022,"version":"3.51.4"},"reference-count":73,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["2134037"],"award-info":[{"award-number":["2134037"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Inform. Theory"],"published-print":{"date-parts":[[2026,5]]},"DOI":"10.1109\/tit.2026.3676355","type":"journal-article","created":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T20:06:23Z","timestamp":1774296383000},"page":"3339-3358","source":"Crossref","is-referenced-by-count":0,"title":["Local Flow Matching Generative Models"],"prefix":"10.1109","volume":"72","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6623-7084","authenticated-orcid":false,"given":"Chen","family":"Xu","sequence":"first","affiliation":[{"name":"H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, GA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1034-6019","authenticated-orcid":false,"given":"Xiuyuan","family":"Cheng","sequence":"additional","affiliation":[{"name":"Department of Mathematics, Duke University, Durham, NC, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6777-2951","authenticated-orcid":false,"given":"Yao","family":"Xie","sequence":"additional","affiliation":[{"name":"H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, GA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"2672","article-title":"Generative adversarial nets","volume-title":"Proc. NeurIPS","volume":"27","author":"Goodfellow"},{"key":"ref2","article-title":"Auto-encoding variational Bayes","volume-title":"Proc. ICLR","author":"Kingma"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.2992934"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3446374"},{"key":"ref5","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume-title":"Proc. NeurIPS","volume":"33","author":"Ho"},{"key":"ref6","article-title":"Score-based generative modeling through stochastic differential equations","volume-title":"Proc. ICLR","author":"Song"},{"key":"ref7","article-title":"Flow matching for generative modeling","volume-title":"Proc. ICLR","author":"Lipman"},{"key":"ref8","article-title":"Building normalizing flows with stochastic interpolants","volume-title":"Proc. ICLR","author":"Albergo"},{"key":"ref9","article-title":"Stochastic interpolants: A unifying framework for flows and diffusions","author":"Albergo","year":"2023","journal-title":"arXiv:2303.08797"},{"key":"ref10","first-page":"6185","article-title":"Variational Wasserstein gradient flow","volume-title":"Proc. ICML","author":"Fan"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2051"},{"key":"ref12","article-title":"DiffWave: A versatile diffusion model for audio synthesis","volume-title":"Proc. ICLR","author":"Kong"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01042"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.026"},{"key":"ref15","article-title":"Rectified flow: A marginal preserving approach to optimal transport","author":"Liu","year":"2022","journal-title":"arXiv:2209.14577"},{"key":"ref16","first-page":"12606","article-title":"Scaling rectified flow transformers for high-resolution image synthesis","volume-title":"Proc. ICML","author":"Esser"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/Humanoids58906.2024.10769838"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1848"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.52202\/079017-4239"},{"key":"ref20","article-title":"FFJORD: Free-form continuous dynamics for scalable reversible generative models","volume-title":"Proc. ICLR","author":"Grathwohl"},{"key":"ref21","article-title":"Progressive distillation for fast sampling of diffusion models","volume-title":"Proc. ICLR","author":"Salimans"},{"key":"ref22","article-title":"Flow straight and fast: Learning to generate and transfer data with rectified flow","volume-title":"Proc. ICLR","author":"Liu"},{"key":"ref23","first-page":"32211","article-title":"Consistency models","volume-title":"Proc. ICML","author":"Song"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2019.2924428"},{"key":"ref25","first-page":"15243","article-title":"Large-scale Wasserstein gradient flows","volume-title":"Proc. NeurIPS","volume":"34","author":"Mokrov"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-31521-y"},{"key":"ref27","first-page":"6571","article-title":"Neural ordinary differential equations","volume-title":"Proc. NeurIPS","volume":"31","author":"Chen"},{"key":"ref28","first-page":"3154","article-title":"How to train your neural ODE: The world of Jacobian and kinetic regularization","volume-title":"Proc. ICML","author":"Finlay"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i10.17113"},{"key":"ref30","first-page":"2995","article-title":"Normalizing Kalman filters for multivariate time series analysis","volume-title":"Proc. NeurIPS","volume":"33","author":"B\u00e9zenac"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i6.25834"},{"key":"ref32","first-page":"2872","article-title":"Computing high-dimensional optimal transport by flow neural networks","volume-title":"Proc. 28th Int. Conf. Artif. Intell. Statist.","author":"Chen"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.130.171402"},{"key":"ref34","article-title":"Improving and generalizing flow-based generative models with minibatch optimal transport","volume":"2024","author":"Tong","year":"2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref35","article-title":"Mean flows for one-step generative modeling","author":"Geng","year":"2025","journal-title":"arXiv:2505.13447"},{"key":"ref36","article-title":"Distilling the knowledge from conditional normalizing flows","volume-title":"Proc. ICML Workshop Invertible Neural Netw.","author":"Baranchuk"},{"key":"ref37","article-title":"InstaFlow: One step is enough for high-quality diffusion-based text-to-image generation","volume-title":"Proc. ICLR","author":"Liu"},{"key":"ref38","article-title":"Bellman optimal stepsize straightening of flow-matching models","volume-title":"Proc. ICLR","author":"Nguyen"},{"key":"ref39","first-page":"946","article-title":"Convergence of score-based generative modeling for general data distributions","volume-title":"Proc. Int. Conf. Algorithmic Learn. Theory","author":"Lee"},{"key":"ref40","article-title":"Sampling is as easy as learning the score: Theory for diffusion models with minimal data assumptions","volume-title":"Proc. ICLR","author":"Chen"},{"key":"ref41","first-page":"4735","article-title":"Improved analysis of score-based generative modeling: User-friendly bounds under minimal smoothness assumptions","volume-title":"Proc. ICML","author":"Chen"},{"key":"ref42","article-title":"Nearly d-linear convergence bounds for diffusion models via stochastic localization","volume-title":"Proc. ICLR","author":"Benton"},{"key":"ref43","first-page":"4462","article-title":"Restoration-degradation beyond linear diffusions: A non-asymptotic analysis for DDIM-type samplers","volume-title":"Proc. ICML","author":"Chen"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2998"},{"key":"ref45","article-title":"Towards non-asymptotic convergence for diffusion-based generative models","volume-title":"Proc. ICLR","author":"Li"},{"key":"ref46","article-title":"A sharp convergence theory for the probability flow ODEs of diffusion models","author":"Li","year":"2024","journal-title":"arXiv:2408.02320"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2025.3557050"},{"key":"ref48","first-page":"1","article-title":"Distribution learning via neural differential equations: A nonparametric statistical perspective","author":"Marzouk","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref49","article-title":"Error bounds for flow matching methods","volume":"2024","author":"Benton","year":"2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref50","article-title":"Convergence of continuous normalizing flows for learning probability distributions","author":"Gao","year":"2024","journal-title":"arXiv:2404.00551"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.52202\/079017-4393"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2024.3422412"},{"key":"ref53","volume-title":"Gradient Flows: In Metric Spaces and in the Space of Probability Measures","author":"Ambrosio","year":"2005"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1137\/S0036141096303359"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1016\/j.jfa.2012.07.007"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2016.2549542"},{"key":"ref57","first-page":"6626","article-title":"GANs trained by a two time-scale update rule converge to a local Nash equilibrium","volume-title":"Proc. NeurIPS","volume":"30","author":"Heusel"},{"key":"ref58","first-page":"2338","article-title":"Masked autoregressive flow for density estimation","volume-title":"Proc. NeurIPS","volume":"30","author":"Papamakarios"},{"key":"ref59","first-page":"1","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/ICVGIP.2008.47"},{"key":"ref62","article-title":"LSUN: Construction of a large-scale image dataset using deep learning with humans in the loop","author":"Yu","year":"2015","journal-title":"arXiv:1506.03365"},{"key":"ref63","article-title":"What matters in learning from offline human demonstrations for robot manipulation","author":"Mandlekar","year":"2021","journal-title":"arXiv:2108.03298"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.52202\/079017-4405"},{"key":"ref65","article-title":"Convex potential flows: Universal probability distributions with optimal transport and convex optimization","volume-title":"Proc. ICLR","author":"Huang"},{"key":"ref66","first-page":"1263","article-title":"Block neural autoregressive flow","volume-title":"Proc. UAI","author":"Cao"},{"key":"ref67","first-page":"22863","article-title":"A variational perspective on diffusion-based generative models and score matching","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","author":"Huang"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2101344118"},{"key":"ref69","first-page":"1188","article-title":"AdaCat: Adaptive categorical discretization for autoregressive models","volume-title":"Proc. 38th Conf. Uncertainty Artif. Intell.","author":"Li"},{"key":"ref70","article-title":"Marginalizable density models","author":"Gilboa","year":"2021","journal-title":"arXiv:2106.04741"},{"key":"ref71","first-page":"8162","article-title":"Improved denoising diffusion probabilistic models","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","volume":"139","author":"Nichol"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01112"},{"key":"ref73","first-page":"9902","article-title":"Planning with diffusion for flexible behavior synthesis","volume-title":"Proc. ICML","author":"J\u00e4nner"}],"container-title":["IEEE Transactions on Information Theory"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/18\/11493629\/11450459-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/18\/11493629\/11450459.pdf?arnumber=11450459","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T19:58:37Z","timestamp":1776974317000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11450459\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":73,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tit.2026.3676355","relation":{},"ISSN":["0018-9448","1557-9654"],"issn-type":[{"value":"0018-9448","type":"print"},{"value":"1557-9654","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]}}}