Data to 7 Oct 2026

Generalized Denoising Auto-Encoders as Generative Models

arXiv 1305.6663 ↗First posted 29 May 2013cs.LGcited by 113

Not on the map. Why?

Cited by 21 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Recent work has shown how denoising and contractive autoencoders implicitly capture the structure of the data-generating density, in the case where the corruption noise is Gaussian, the reconstruction error is the squared error, and the data is continuous-valued. This has led to various proposals for sampling from this implicitly learned density function, using Langevin and Metropolis-Hastings MCMC. However, it remained unclear how to connect the training procedure of regularized auto-encoders to the implicit estimation of the underlying data-generating distribution when the data are discrete, or using other forms of corruption process and reconstruction errors. Another issue is the mathematical justification which is only valid in the limit of small corruption noise. We propose here a different attack on the problem, which deals with all these issues: arbitrary (but noisy enough) corruption, arbitrary reconstruction loss (seen as a log-likelihood), handling both discrete and continuous-valued variables, and removing the bias due to non-infinitesimal corruption noise (or non-infinitesimal contractive penalty).

Key facts. Cited by 113 papers in the map (data through 7 Oct 2026). First posted on arXiv 29 May 2013.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 17; 2022: 21; 2023: 17; 2024: 15; 2025: 22; 2026: 21.

113 papers.

Sort:

By area:

2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
2021Score-based Generative Modeling in Latent Spacecited by 634
2022A Unified Model for Multi-class Anomaly Detectioncited by 267
2021Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Modelscited by 251
2021Spatio-temporal Self-Supervised Representation Learning for 3D Point Cloudscited by 180
2022Multi-Objective GFlowNetscited by 100
2023Diffusion Model as Representation Learnercited by 95
2021Learning Neural Generative Dynamics for Molecular Conformation Generationcited by 87
2023Deep Learning for Time Series Classification and Extrinsic Regression: A Current Surveycited by 68
2022Latent Diffusion Energy-Based Model for Interpretable Text Modelingcited by 67
2022Masked Vision and Language Modeling for Multi-modal Representation Learningcited by 66
2021Generative Zero-Shot Learning for Semantic Segmentation of 3D Point Cloudscited by 55

38 more

2024Diffusion Models and Representation Learning: A Surveycited by 53
2024On the Trajectory Regularity of ODE-based Diffusion Samplingcited by 51
2021MIRACLE: Causally-Aware Imputation via Learning Missing Data Mechanismscited by 50
2022On Analyzing Generative and Denoising Capabilities of Diffusion-based Deep Generative Modelscited by 31
2025AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detectioncited by 28
2025MC3D-AD: A Unified Geometry-aware Reconstruction Model for Multi-category 3D Anomaly Detectioncited by 23
2023Blind Audio Bandwidth Extension: A Diffusion-Based Zero-Shot Approachcited by 21
2023Learning Energy-Based Models by Cooperative Diffusion Recovery Likelihoodcited by 19
2023Embedding in Recommender Systems: A Surveycited by 19
2025Learning to Detect Multi-class Anomalies with Just One Normal Image Promptcited by 18
2023A Geometric Perspective on Diffusion Modelscited by 17
2023Energy-Based Models for Anomaly Detection: A Manifold Diffusion Recovery Approachcited by 17
2022POS-BERT: Point Cloud One-Stage BERT Pre-Trainingcited by 16
2022ZigZag: Universal Sampling-free Uncertainty Estimation Through Two-Step Inferencecited by 15
2026Self-Refining Video Samplingcited by 14
2024Visual Lexicon: Rich Image Features in Language Spacecited by 12
2025MetaUAS: Universal Anomaly Segmentation with One-Prompt Meta-Learningcited by 12
2022Probing BERT's priors with serial reproduction chainscited by 11
2024Enabling Uncertainty Estimation in Iterative Neural Networkscited by 11
2021Model Selection for Bayesian Autoencoderscited by 10
2024Distributional Principal Autoencoderscited by 10
2025Generalization of Diffusion Models Arises with a Balanced Representation Spacecited by 10
2026VisualAD: Language-Free Zero-Shot Anomaly Detection via Vision Transformercited by 10
2023Training Energy-Based Models with Diffusion Contrastive Divergencescited by 9
2022A composable machine-learning approach for steady-state simulations on high-resolution gridscited by 8
2023Deriving Language Models from Masked Language Modelscited by 8
2024Exploring the design space of deep-learning-based weather forecasting systemscited by 8
2023Moment Matching Denoising Gibbs Samplingcited by 7
2025Flow Matching for Collaborative Filteringcited by 7
2025Inference-Time Scaling of Diffusion Language Models via Trajectory Refinementcited by 7
2021Noise Learning Based Denoising Autoencodercited by 6
2024Real-Time Zero-Day Intrusion Detection System for Automotive Controller Area Network on FPGAscited by 5
2024FiffDepth: Feed-forward Transformation of Diffusion-Based Generators for Detailed Depth Estimationcited by 5
2025Are We Really Learning the Score Function? Reinterpreting Diffusion Models Through Wasserstein Gradient Flow Matchingcited by 5
2021On the Generative Utility of Cyclic Conditionalscited by 4
2023DiffPMAE: Diffusion Masked Autoencoders for Point Cloud Reconstructioncited by 4
2024SQUWA: Signal Quality Aware DNN Architecture for Enhanced Accuracy in Atrial Fibrillation Detection from Noisy PPG Signalscited by 4
2024Robust Generative Learning with Lipschitz-Regularized α-Divergences Allows Minimal Assumptions on Target Distributionscited by 4

50 most cited of 113 All 113 (JSON)