Your Absorbing Discrete Diffusion Secretly Models the Bayesian Posterior

• URL: http://arxiv.org/abs/2507.07586v2
• Date: Sun, 13 Jul 2025 12:37:30 GMT
• Title: Your Absorbing Discrete Diffusion Secretly Models the Bayesian Posterior
• Abstract summary: We prove that the expected denoiser output under the forward corruption distribution recovers the true posterior.<n>We introduce an inference-time ensemble that runs K independent denoising passes and aggregates both posterior means and variances.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Discrete diffusion language models learn to reconstruct text from randomly masked inputs, yet under mild assumptions their denoiser already implements the exact Bayesian posterior over the original tokens. We prove that the expected denoiser output under the forward corruption distribution recovers the true posterior, and that a simple Monte Carlo estimator converges to this posterior at rate O(1/sqrt(K)) with finite-sample concentration bounds. Building on this insight, we introduce an inference-time ensemble that runs K independent denoising passes and aggregates both posterior means and variances without any extra training. On WikiText-2, our MC-marginal sampler recovers the analytic lambda-DCE zero-shot perplexity (approximately 39) to within a few points at K=128, and its per-token variance shows a strong rank correlation with reconstruction error (Spearman rho = 0.996). This cost-proportional procedure yields calibrated uncertainty estimates and a direct trade-off between compute and posterior fidelity in discrete diffusion LMs.

Related papers

• Impute-EM: Native Mixed-State Diffusion Models for Heterogeneous Data Imputation [71.74203645336385]
  Impute-EM alternates between imputing missing entries with the current model and refitting a diffusion backbone on completed data.<n>We characterize the update and show that the observed mask-indexed marginals match the targets at the limit.
  arXiv  Detail & Related papers  (2026-09-14T09:39:13Z)
• Scale-Consistent Posterior Dynamics for Diffusion Inverse Problems [16.685352002333225]
  Posterior sampling with a pretrained diffusion prior is governed by a conditional score whose likelihood component is generally intractable.<n>We show that a noise-conditioned path whose targets approach the clean posterior produces a continuous surrogate SDE.<n>Experiments on FFHQ and ImageNet with 100 score evaluations demonstrate competitive reconstruction fidelity for superresolution and deblurring.
  arXiv  Detail & Related papers  (2026-08-15T09:38:41Z)
• A Unified Kullback--Leibler Divergence Analysis of Generative Diffusion Models via Entropy Production Rate [4.625787677472786]
  We introduce a unified framework for the error analysis of generative models based on the entropy production rate of the forward-reverse diffusion process pair.<n>By analyzing the entropy production at the level of marginal distributions, rather than in path space, our framework yields a sharp convergence rate of $mathcalO(h2)$ for the Euler-Maruyama sampler.
  arXiv  Detail & Related papers  (2026-08-03T15:49:08Z)
• Uniform Diffusion Models Revisited: Leave-One-Out Denoiser and Absorbing State Reformulation [72.05592785529404]
  We show that the standard plug-in bridge parameterization for UDM is not optimized by the denoising posterior.<n>We also introduce an absorbing-state reformulation of uniform diffusion that preserves the UDM joint law while decomposing it into masked-diffusion-like sampling operations.
  arXiv  Detail & Related papers  (2026-05-21T17:27:19Z)
• Sharp Convergence Rates for Masked Diffusion Models [53.117058231393834]
  We develop a total-variation based analysis for the Euler method that overcomes limitations.<n>Our results relax assumptions on score estimation, improve parameter dependencies, and establish convergence guarantees.<n>Overall, our analysis introduces a direct TV-based error decomposition along the CTMC trajectory and a decoupling-based path-wise analysis for FHS.
  arXiv  Detail & Related papers  (2026-02-26T00:47:51Z)
• Nearly Optimal Bayesian Inference for Structural Missingness [23.988531482641307]
  In the Bayesian view, prediction via the posterior predictive distribution integrates over the full model posterior uncertainty.<n>This framework decouples learning an in-model missing-value posterior from (ii) label prediction by optimizing the predictive posterior distribution.<n>It achieves SOTA on 43 classification and 15 imputation benchmarks, with finite-sample near Bayes-optimality guarantees.
  arXiv  Detail & Related papers  (2026-01-26T14:03:11Z)
• Provable Diffusion Posterior Sampling for Bayesian Inversion [13.807494493914335]
  This paper proposes a novel diffusion-based posterior sampling method within a plug-and-play framework.<n>To approximate the posterior score, we develop a Monte Carlo estimator in which particles are generated using Langevin dynamics.<n>On the theoretical side, we provide non-asymptotic error bounds, showing that the method converges even for complex multi-modal target posterior.
  arXiv  Detail & Related papers  (2025-12-08T20:34:05Z)
• Antithetic Noise in Diffusion Models [13.216777115252563]
  We find that pairing each initial noise with its negation consistently yields strongly negatively correlated samples.<n>Our framework is training-free, model-agnostic, and adds no runtime overhead.
  arXiv  Detail & Related papers  (2025-06-06T15:46:26Z)
• Kernel-Smoothed Scores for Denoising Diffusion: A Bias-Variance Study [3.265950484493743]
  Diffusion models can be prone to memorization.<n>Regularization on the score has the same effect as increasing the size of the training dataset.<n>This perspective highlights two regularization mechanisms taking place in denoising diffusions.
  arXiv  Detail & Related papers  (2025-05-28T20:22:18Z)
• Robust Conformal Prediction with a Single Binary Certificate [58.450154976190795]
  Conformal prediction (CP) converts any model's output to prediction sets with a guarantee to cover the true label with (adjustable) high probability.<n>We propose a robust conformal prediction that produces smaller sets even with significantly lower MC samples.
  arXiv  Detail & Related papers  (2025-03-07T08:41:53Z)
• Convergence of Score-Based Discrete Diffusion Models: A Discrete-Time Analysis [56.442307356162864]
  We study the theoretical aspects of score-based discrete diffusion models under the Continuous Time Markov Chain (CTMC) framework.<n>We introduce a discrete-time sampling algorithm in the general state space $[S]d$ that utilizes score estimators at predefined time points.<n>Our convergence analysis employs a Girsanov-based method and establishes key properties of the discrete score function.
  arXiv  Detail & Related papers  (2024-10-03T09:07:13Z)
• Epistemic Uncertainty and Observation Noise with the Neural Tangent Kernel [12.464924018243988]
  Recent work has shown that training wide neural networks with gradient descent is formally equivalent to computing the mean of the posterior distribution in a Gaussian Process. We show how to deal with non-zero aleatoric noise and derive an estimator for the posterior covariance.
  arXiv  Detail & Related papers  (2024-09-06T00:34:44Z)
• Amortizing intractable inference in diffusion models for vision, language, and control [89.65631572949702]
  This paper studies amortized sampling of the posterior over data, $mathbfxsim prm post(mathbfx)propto p(mathbfx)r(mathbfx)$, in a model that consists of a diffusion generative model prior $p(mathbfx)$ and a black-box constraint or function $r(mathbfx)$.<n>We prove the correctness of a data-free learning objective, relative trajectory balance, for training a diffusion model that samples from
  arXiv  Detail & Related papers  (2024-05-31T16:18:46Z)
• Divide-and-Conquer Posterior Sampling for Denoising Diffusion Priors [21.0128625037708]
  We present an innovative framework, divide-and-conquer posterior sampling. It reduces the approximation error associated with current techniques without the need for retraining. We demonstrate the versatility and effectiveness of our approach for a wide range of Bayesian inverse problems.
  arXiv  Detail & Related papers  (2024-03-18T01:47:24Z)
• Calibrating Neural Simulation-Based Inference with Differentiable Coverage Probability [50.44439018155837]
  We propose to include a calibration term directly into the training objective of the neural model. By introducing a relaxation of the classical formulation of calibration error we enable end-to-end backpropagation. It is directly applicable to existing computational pipelines allowing reliable black-box posterior inference.
  arXiv  Detail & Related papers  (2023-10-20T10:20:45Z)
• Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning [59.44422468242455]
  We propose a novel method dubbed ShrinkMatch to learn uncertain samples. For each uncertain sample, it adaptively seeks a shrunk class space, which merely contains the original top-1 class. We then impose a consistency regularization between a pair of strongly and weakly augmented samples in the shrunk space to strive for discriminative representations.
  arXiv  Detail & Related papers  (2023-08-13T14:05:24Z)
• Unscented Autoencoder [3.0108936184913295]
  Variational Autoencoder (VAE) is a seminal approach in deep generative modeling with latent variables. We apply the Unscented Transform (UT) -- a well-known distribution approximation used in the Unscented Kalman Filter (UKF) from the field of filtering. We derive a novel, deterministic-sampling flavor of the VAE, the Unscented Autoencoder (UAE), trained purely with regularization-like terms on the per-sample posterior.
  arXiv  Detail & Related papers  (2023-06-08T14:53:02Z)
• Bayesian Pseudo-Coresets via Contrastive Divergence [5.479797073162603]
  We introduce a novel approach for constructing pseudo-coresets by utilizing contrastive divergence. It eliminates the need for approximations in the pseudo-coreset construction process. We conduct extensive experiments on multiple datasets, demonstrating its superiority over existing BPC techniques.
  arXiv  Detail & Related papers  (2023-03-20T17:13:50Z)
• Regularized Vector Quantization for Tokenized Image Synthesis [126.96880843754066]
  Quantizing images into discrete representations has been a fundamental problem in unified generative modeling. deterministic quantization suffers from severe codebook collapse and misalignment with inference stage while quantization suffers from low codebook utilization and reconstruction objective. This paper presents a regularized vector quantization framework that allows to mitigate perturbed above issues effectively by applying regularization from two perspectives.
  arXiv  Detail & Related papers  (2023-03-11T15:20:54Z)
• Do Bayesian Variational Autoencoders Know What They Don't Know? [0.6091702876917279]
  The problem of detecting the Out-of-Distribution (OoD) inputs is paramount importance for Deep Neural Networks. It has been previously shown that even Deep Generative Models that allow estimating the density of the inputs may not be reliable. This paper investigates three approaches to inference: Markov chain Monte Carlo, Bayes gradient by Backpropagation and Weight Averaging-Gaussian.
  arXiv  Detail & Related papers  (2022-12-29T11:48:01Z)
• Score-based Continuous-time Discrete Diffusion Models [102.65769839899315]
  We extend diffusion models to discrete variables by introducing a Markov jump process where the reverse process denoises via a continuous-time Markov chain. We show that an unbiased estimator can be obtained via simple matching the conditional marginal distributions. We demonstrate the effectiveness of the proposed method on a set of synthetic and real-world music and image benchmarks.
  arXiv  Detail & Related papers  (2022-11-30T05:33:29Z)
• Multi-hypothesis 3D human pose estimation metrics favor miscalibrated distributions [3.8575800313102806]
  Well-calibrated distributions can make ambiguities explicit and preserve uncertainty for downstream tasks. We identify that miscalibration can be attributed to the use of sample-based metrics such as minMPJPE. To mitigate this problem, we propose an accurate and well-calibrated model called Conditional Graph Normalizing Flow (cGNF)
  arXiv  Detail & Related papers  (2022-10-20T11:47:07Z)
• Robust Implicit Networks via Non-Euclidean Contractions [63.91638306025768]
  Implicit neural networks show improved accuracy and significant reduction in memory consumption. They can suffer from ill-posedness and convergence instability. This paper provides a new framework to design well-posed and robust implicit neural networks.
  arXiv  Detail & Related papers  (2021-06-06T18:05:02Z)
• Bayesian Uncertainty Estimation of Learned Variational MRI Reconstruction [63.202627467245584]
  We introduce a Bayesian variational framework to quantify the model-immanent (epistemic) uncertainty. We demonstrate that our approach yields competitive results for undersampled MRI reconstruction.
  arXiv  Detail & Related papers  (2021-02-12T18:08:14Z)
• Multi-label Contrastive Predictive Coding [125.03510235962095]
  Variational mutual information (MI) estimators are widely used in unsupervised representation learning methods such as contrastive predictive coding (CPC) We introduce a novel estimator based on a multi-label classification problem, where the critic needs to jointly identify multiple positive samples at the same time. We show that using the same amount of negative samples, multi-label CPC is able to exceed the $log m$ bound, while still being a valid lower bound of mutual information.
  arXiv  Detail & Related papers  (2020-07-20T02:46:21Z)
• Consistency Regularization for Certified Robustness of Smoothed Classifiers [89.72878906950208]
  A recent technique of randomized smoothing has shown that the worst-case $ell$-robustness can be transformed into the average-case robustness. We found that the trade-off between accuracy and certified robustness of smoothed classifiers can be greatly controlled by simply regularizing the prediction consistency over noise.
  arXiv  Detail & Related papers  (2020-06-07T06:57:43Z)
• A Batch Normalized Inference Network Keeps the KL Vanishing Away [35.40781000297285]
  Variational Autoencoder (VAE) is widely used to approximate a model's posterior on latent variables. VAE often converges to a degenerated local optimum known as "posterior collapse"
  arXiv  Detail & Related papers  (2020-04-27T05:20:01Z)
• Generative Modeling with Denoising Auto-Encoders and Langevin Sampling [88.83704353627554]
  We show that both DAE and DSM provide estimates of the score of the smoothed population density. We then apply our results to the homotopy method of arXiv:1907.05600 and provide theoretical justification for its empirical success.
  arXiv  Detail & Related papers  (2020-01-31T23:50:03Z)

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.