Score-Based Generative Modeling

Follow this topic Ask about this topic

Momentum

6 papers in the last four weeks, with none the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 71

All topics

Sample complexity bounds for categorical Markov random fields via Discrete Diffusions
Oct 1, 2026Shivam Kumar, Nabarun DebMarkov Random Fields Diffusion Sampling
Booth School of Business, University of Chicago
Weighting Schedules Govern What and When Score-Based Generative Models Learn from Multimodal Data
Sep 28, 2026Jérémie Klinger, Raphaël Urfin, Giulio Biroli +1Generative Modeling Score-Based Generative Modeling
Laboratoire de Physique de l’École normale supérieure, ENS, Université PSL, CNRS, Sorbonne Université, Université Paris Cité, F-75005 Paris, France
Quantum score matching with applications to learning thermal states
Sep 23, 2026Yulong Dong, Jiaqi LengQuantum Machine Learning Score-Based Generative Modeling
Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, Michigan 48109, USA · Institute for Advanced Computing, Virginia Tech, Alexandria, Virginia 22305, USA · Department of Computer Science, Virginia Tech, Alexandria, Virginia 22305, USA
Robustness of Diffusion Models under Distribution Shift
Sep 23, 2026Wei Luo, Neil K. Chada, Shijie Zhang +1Distribution Shift Robustness Diffusion Sampling
Department of Mathematical Sciences, Tsinghua University · Department of Mathematics, City University of Hong Kong, Hong Kong SAR · Artificial Intelligence Research Institute, Shenzhen MSU-BIT University +1
Walking the Score Manifold: Continuous-time Generative Dynamics on Learned Data Manifolds
Sep 15, 2026Jan Tauberschmidt, Brian B. Moser, Stanislav Frolov +3Dynamical Systems Latent Dynamics Modeling
DFKI Kaiserslautern · RPTU Kaiserslautern · Imperial College London
Score-based Outlier Generation via Controlling the Radon-Nikodym Derivative
Sep 14, 2026Amartya Mukherjee, Tristan Milne, Kry Yik-Chau Lui +2Diffusion Sampling Synthetic Anomaly Generation
Department of Applied Mathematics, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 · Royal Bank of Canada, 1 Place Ville Marie, Montreal, Quebec, H3C 3A9
Intrinsic-Hybrid Latent Diffusion Models for Generative Modeling on Unknown Manifolds
Aug 5, 2026Yizhu Wang, Mu Niu, Xiaochen YangLatent Diffusion Models Generative Modeling
School of Mathematics and Statistics, University of Glasgow, Glasgow, U.K.
A Hyperfinite Framework for Score-Based Generative Modeling
Aug 3, 2026Sunder Ram KrishnanScore Matching Stochastic Differential Equations
Kerala School of Mathematics, Kozhikode 673571, Kerala, India.
Conditioning Tree-Based Diffusions and Flows for Probabilistic Tabular Regression
Jul 30, 2026Silas KoemenFlow Matching Gradient-Boosted Decision Trees
Independent Researcher
From Score Learning to Discretized Sampling: An End-to-End Generalization Analysis of Diffusion Models
Jul 25, 2026Jinshu Huang, Yiming Jiang, Chunlin WuDiffusion Model Sampling Score-Based Generative Modeling
School of Mathematical Sciences, Nankai University, Tianjin, China
From Score Approximation to Distribution Approximation in Score-Based Diffusion Models
Jul 24, 2026Lan V. TruongScore-Based Generative Modeling Diffusion Models
Faculty of Computer Science and Engineering · Ho Chi Minh City University of Technology (HCMUT) · Vietnam National University Ho Chi Minh City (VNU-HCM), Vietnam
Toward a mechanistic understanding of inference in visual cortex and diffusion models
Jul 17, 2026Zeyu Yun, Alexander Belsten, Dasheng Bi +3Mechanistic Interpretability Visual Neuroscience
Dept. of Electrical Engineering and Computer Science, UC Berkeley · Redwood Center for Theoretical Neuroscience, UC Berkeley · Herbert Wertheim School of Optometry and Vision Science, UC Berkeley +3
Quantum Circuits in Diffusion Models: A Fair-Comparison Study and a Mechanistic Analysis of Angle-Embedding Failures
Jul 10, 2026Jaeuk Kim, Sanghoon YooQuantum Generative Modeling Variational Quantum Circuits
NextITS Co., Ltd.
Tensor Train Diffusion: Leveraging Low-Rank Structures for High-Dimensional Score-Based Sampling
Jul 7, 2026Robert Gruhlke, Julius Berner, David Sommer +1Stochastic Differential Equations Diffusion Sampling
Freie Universität Berlin · NVIDIA · WIAS +2
Non-Asymptotic Error Bounds for SMC with Biased Proposals: Application to Conditional Diffusion Sampling
Jul 6, 2026Stanislas Strasman, Gabriel Victorino Cardoso, Sylvain Le Corff +2Markov Chain Monte Carlo Markov Models
SU, LPSM · Sorbonne Université and Université Paris Cité, CNRS, LPSM, F-75005 Paris, France · LPSM +2
Non-asymptotic Convergence of Stochastic Gradient Descent in Score-based Generative Models
Jul 6, 2026Stanislas Strasman, Sobihan Surendran, Sylvain Le CorffNonconvex Stochastic Optimization Score-Based Generative Modeling
SU, LPSM
Tightening the Score Matching Gap for Diffusion Models
Jul 5, 2026Benjamin Dupuis, Tyler Farghly, Maxime Haddouche +2Score Matching Score-Based Generative Modeling
INRIA - CNRS - D´epartement d’Informatique de l’Ecole Normale Sup´erieure - PSL Research University, France · Ecole Polytechnique - CMAP, IP Paris, Palaiseau, France
Class-frequency Guided Noise Schedule for Diffusion Models
Jun 26, 2026Jiequan Cui, Beier Zhu, Qingshan Xu +3Image Generation Score-Based Generative Modeling
Hefei University of Technology. · University of Science and Technology of China. · The University of Hong Kong. +2
Sampling the Schwinger Model with Gauge-Equivariant Diffusion
Jun 25, 2026Octavio Vega, Aida X. El-KhadraDiffusion Sampling Generative Modeling
Department of Physics University of Illinois Urbana-Champaign Urbana, IL 61801
LISA: Likelihood Score Alignment for Visual-condition Controllable Generation
Jun 25, 2026Yanghao Wang, Hongxu Chen, Jiazhen Liu +4Score-Based Generative Modeling Controllable Video Generation
The Hong Kong University of Science and Technology · Huawei Research
Laplace-Fisher Gate Identities for Optimal Matrix-Gated Blended Score Estimation
Jun 23, 2026Alois Duston, Tan Bui-ThanhBayesian Inference Score-Based Generative Modeling
The Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin · The Oden Institute for Computational Engineering and Sciences, and the Department of Aerospace Engineering & Engineering Mechanics, The University of Texas at Austin
$Ω$ : Operator-based Mixture Ensemble for Generative Assimilation
Jun 18, 2026Pouria Behnoudfar, Nan ChenData Assimilation Posterior Sampling
University of Wisconsin-Madison, Department of Mathematics, Madison, WI, USA
Score Approximation for Diffusion Models on Arbitrary Low-Dimensional Structures
Jun 18, 2026Xinhe Mu, Zaijiu Shang, Zhaoqi Zhou +4Neural Network Approximation Theory Score-Based Generative Modeling
Academy of Mathematics and Systems Sciences Chinese Academy of Sciences Beijing, China · Shanghai Institute for Mathematics and Interdisciplinary Sciences Shanghai, China · Huawei Technologies Co., Ltd. Beijing, China
Volterra Generative Models
Jun 16, 2026Yusen Jia, Bingyan HanDiffusion Sampling Diffusion-Based Image Generation
The Hong Kong University of Science and Technology (Guangzhou)
Exact Posterior Score Estimation for Solving Linear Inverse Problems
Jun 15, 2026Abbas Mammadov, Ozgur Kara, Kaan Oktay +5Diffusion Models for Image Restoration Diffusion-Based Inverse Problems
University of Oxford · UIUC · fal +1
A Perception vs. Distortion Perspective on Score-Based Generative Channel Estimation
Jun 15, 2026Marco Skocaj, Lukas Eller, Mate BobanWireless Communications Channel Estimation
All authors are with the Advanced Wireless Technology Laboratory, Huawei Heisenberg Research Centre, Munich, Germany.
When Do Local Score Models Extrapolate Across Size? A Diagnostic Theory and Benchmark
Jun 8, 2026Wenjie XiScore-Based Generative Modeling
Department of Physics and HK Institute of Quantum Science & Technology The University of Hong Kong, Pokfulam Road, Hong Kong, China
Where the Score Lives: A Wavelet View of Diffusion
Jun 6, 2026Emma Finn, Binxu Wang, T. Anderson Keller +1Score-Based Generative Modeling Diffusion Models
The Kempner Instutite for the Study of Natural and Artificial Intelligence Harvard University, Cambridge, MA 15213
Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
Jun 4, 2026Naïl B. Khelifa, Richard E. Turner, Ramji VenkataramananScore Matching Hodge Decomposition
University of Cambridge
Learning Implicit Bias in Generative Spaces for Accelerating Protein Dynamics Emulation
Jun 1, 2026Kaihui Cheng, Zhiqiang Cai, Wenkai Xiang +4Diffusion Model Sampling Adaptive Sampling
1Fudan University · 2Shanghai Academy of AI for Science · 3Shanghai Innovation Institute

Oct 1, 2026, math.STJ/K move · Enter open · S save

Sample complexity bounds for categorical Markov random fields via Discrete Diffusions

Shivam Kumar, Nabarun Deb

Booth School of Business, University of Chicago

Markov Random Fields Diffusion Sampling Score-Based Generative Modeling Diffusion Models

Open paper AskPDF arXiv

Many applications in statistics, economics, and physics require sampling from high-dimensional categorical distributions with local dependence structures. Examples include finite memory language models, Ising and Potts systems in statistical physics and protein folding, etc. In modern machine learning, discrete diffusions have emerged as a flexible approach for sampling such data, with strong empirical performance. Motivated by this, we develop learning methods with end-to-end sample complexity bounds for discrete diffusion with uniform noising under local dependence, which we model through low order Markov random fields (MRFs). Our main technical insight is a new \emph{pinning decomposition} of the discrete score. It shows that unlike in continuous diffusions, the score decomposes into components where the dependence on time separates multiplicatively from the dependence on the target. Building on this decomposition, we propose a \emph{weight-sharing neural score learner} and combine it with

τ

-leaping to obtain an end-to-end sampling procedure. Rather than treating score-learning error as a black-box input, as is common in existing sampling analyses, we study the score learning error from finite data and derive optimal sampling guarantees with explicit dependence on the vocabulary size, the interaction order of the MRF, and the sample size. Moreover, our strategy trains a single score network across uniform noise levels while leaving the sampling discretization to be chosen at inference-time. This allows the same trained model to trade accuracy for computational cost as inference-time budgets vary. Numerical experiments on Potts, Ising, and tree-structured models show that weight-sharing score networks outperform fully connected ones for sampling long sequences.

Momentum

Related topics

Latest papers 71