mlsanityData to 7 Oct 2026

Diffusive Gibbs Sampling

arXiv 2402.03008 ↗First posted 5 Feb 2024stat.ML cs.LG stat.COcited by 21

Not on the map. Why?

Abstract (arXiv)

The inadequate mixing of conventional Markov Chain Monte Carlo (MCMC) methods for multi-modal distributions presents a significant challenge in practical applications such as Bayesian inference and molecular dynamics. Addressing this, we propose Diffusive Gibbs Sampling (DiGS), an innovative family of sampling methods designed for effective sampling from distributions characterized by distant and disconnected modes. DiGS integrates recent developments in diffusion models, leveraging Gaussian convolution to create an auxiliary noisy distribution that bridges isolated modes in the original space and applying Gibbs sampling to alternately draw samples from both spaces. A novel Metropolis-within-Gibbs scheme is proposed to enhance mixing in the denoising sampling step. DiGS exhibits a better mixing property for sampling multi-modal distributions than state-of-the-art methods such as parallel tempering, attaining substantially improved performance across various tasks, including mixtures of Gaussians, Bayesian neural networks and molecular dynamics.

Key facts. Cited by 21 papers in the map (data through 7 Oct 2026). Builds on 28 of its 61 references that are in the map. First posted on arXiv 5 Feb 2024.

Builds on

28 of 61 references.

References · 28

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  5. 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,194
  6. 2020Improved Techniques for Training Score-Based Generative Modelscited by 1,036
  7. 2016Entropy-SGD: Biasing Gradient Descent Into Wide Valleyscited by 357
  8. 2019Rapid Convergence of the Unadjusted Langevin Algorithm: Isoperimetry Sufficescited by 224
  9. 2017Multiplicative Normalizing Flows for Variational Bayesian Neural Networkscited by 157
  10. 2020Stochastic Normalizing Flowscited by 156
  11. 2022Flow Annealed Importance Sampling Bootstrapcited by 124
  12. 2020Learning Energy-Based Models by Diffusion Recovery Likelihoodcited by 122
16 more
  1. 2017Stabilizing Training of Generative Adversarial Networks through Regularizationcited by 118
  2. 2020Structured Logconcave Sampling with a Restricted Gaussian Oraclecited by 85
  3. 2022Improved analysis for a proximal algorithm for samplingcited by 72
  4. 2021Resampling Base Distributions of Normalizing Flowscited by 42
  5. 2023Reverse Diffusion Monte Carlocited by 41
  6. 2020Blindness of score-based methods to isolated components and mixing proportionscited by 33
  7. 2017Learning Energy-Based Models as Generative ConvNets via Multi-grid Modeling and Samplingcited by 32
  8. 2020A Contour Stochastic Gradient Langevin Dynamics Algorithm for Simulations of Multi-modal Distributionscited by 26
  9. 2018Spread Divergencecited by 23
  10. 2016Auxiliary gradient-based sampling algorithmscited by 21
  11. 2018A Framework for Adaptive MCMC Targeting Multimodal Distributionscited by 21
  12. 2022Towards Healing the Blindness of Score Matchingcited by 13
  13. 2022Non-reversible Parallel Tempering for Deep Posterior Approximationcited by 7
  14. 2023Moment Matching Denoising Gibbs Samplingcited by 7
  15. 2023Entropy-MCMC: Sampling from Flat Basins with Easecited by 6
  16. 2021Spread Flows for Manifold Modellingcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 4; 2025: 10; 2026: 7.

21 papers.

Sort:

By area:

  1. 2024Training Neural Samplers with Reverse Diffusive KL Divergencecited by 21
  2. 2024BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matchingcited by 14
  3. 2024Improving Probabilistic Diffusion Models With Optimal Diagonal Covariance Matchingcited by 13
  4. 2025Progressive Tempering Sampler with Diffusioncited by 13
  5. 2024Bayesian Experimental Design via Contrastive Diffusionscited by 9
  6. 2025CREPE: Controlling Diffusion with Replica Exchangecited by 9
  7. 2025Rethinking Losses for Diffusion Bridge Samplerscited by 7
  8. 2025Inference-Time Scaling of Diffusion Language Models via Trajectory Refinementcited by 7
  9. 2025End-To-End Learning of Gaussian Mixture Priors for Diffusion Samplercited by 4
  10. 2025Geodesic Slice Sampler for Multimodal Distributions with Strong Curvaturecited by 2
  11. 2026Diffusion Path Samplers via Sequential Monte Carlocited by 2
  12. 2025Sampling by averaging: A multiscale approach to score estimationcited by 1
9 more
  1. 2026Alternating Diffusion for Proximal Sampling with Zeroth Order Queriescited by 1
  2. 2026Conditional Diffusion Samplingcited by 1
  3. 2026Exact Global MCMC with Denoising Diffusioncited by 1
  4. 2025Potential Score Matching: Debiasing Molecular Structure Sampling with Potential Energy Guidancecited by 0
  5. 2025Feature Preserving Shrinkage on Bayesian Neural Networks via the R2D2 Priorcited by 0
  6. 2025LSAM: Asynchronous Distributed Training with Landscape-Smoothed Sharpness-Aware Minimizationcited by 0
  7. 2026Sampling via Stochastic Interpolants by Langevin-based Velocity and Initialization Estimation in Flow ODEscited by 0
  8. 2026Entropic Mirror Monte Carlocited by 0
  9. 2026Slithering Through Gaps: Capturing Discrete Isolated Modes via Logistic Bridgingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Interleaved Gibbs Diffusion: Generating Discrete-Continuous Data with Implicit Constraints
  2. 2026From Scores to Gibbs Correctors: Accelerating Uniform-Rate Discrete Diffusion Models
  3. 2023Accelerating Markov Chain Monte Carlo sampling with diffusion models
  4. 2026Markov Chain Monte Carlo with Diffusion Paths
  5. 2026Composing diffusion priors with explicit physical context via generative Gibbs sampling
  6. 2025Progressive Inference-Time Annealing of Diffusion Models for Sampling from Boltzmann Densities
  7. 2026Learning to Replace MCMC in Split-Gibbs Diffusion Posterior Sampling via Deep Unfolding
  8. 2025Split Gibbs Discrete Diffusion Posterior Sampling