mlsanityData to 7 Oct 2026

Robust Classification by Coupling Data Mollification with Label Smoothing

arXiv 2406.01494 ↗First posted 3 Jun 2024cs.CV cs.LG stat.MLcited by 2

Not on the map. Why?

Abstract (arXiv)

Introducing training-time augmentations is a key technique to enhance generalization and prepare deep neural networks against test-time corruptions. Inspired by the success of generative diffusion models, we propose a novel approach of coupling data mollification, in the form of image noising and blurring, with label smoothing to align predicted label confidences with image degradation. The method is simple to implement, introduces negligible overheads, and can be combined with existing augmentations. We demonstrate improved robustness and uncertainty quantification on the corrupted image benchmarks of CIFAR, TinyImageNet and ImageNet datasets.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 49 of its 73 references that are in the map. First posted on arXiv 3 Jun 2024.

Builds on

49 of 73 references.

References · 49

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  5. 2022Classifier-Free Diffusion Guidancecited by 5,717
  6. 2022Generative Adversarial Networkscited by 5,603
  7. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  8. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  9. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  10. 2017On Calibration of Modern Neural Networkscited by 4,860
  11. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  12. 2016Wide Residual Networkscited by 3,057
37 more
  1. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  2. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  3. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  4. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  5. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  6. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  7. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
  8. 2021Variational Diffusion Modelscited by 1,145
  9. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
  10. 2018Evidential Deep Learning to Quantify Classification Uncertaintycited by 728
  11. 2015Importance Weighted Autoencoderscited by 454
  12. 2020Improving robustness against common corruptions by covariate shift adaptationcited by 427
  13. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 376
  14. 2021ResNet strikes back: An improved training procedure in timmcited by 365
  15. 2021What Are Bayesian Neural Network Posteriors Really Like?cited by 307
  16. 2019On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networkscited by 304
  17. 2023Your Diffusion Model is Secretly a Zero-Shot Classifiercited by 301
  18. 2022Cold Diffusion: Inverting Arbitrary Image Transforms Without Noisecited by 275
  19. 2021TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentationcited by 225
  20. 2022Improving Out-of-Distribution Robustness via Selective Augmentationcited by 202
  21. 2022Generative Modelling With Inverse Heat Dissipationcited by 164
  22. 2021PixMix: Dreamlike Pictures Comprehensively Improve Safety Measurescited by 131
  23. 2022Blurring Diffusion Modelscited by 89
  24. 2022On Uncertainty, Tempering, and Data Augmentation in Bayesian Classificationcited by 46
  25. 2021Martingale posterior distributionscited by 43
  26. 2021Data augmentation in Bayesian neural networks and the cold posterior effectcited by 36
  27. 2020Regularization via Structural Label Smoothingcited by 29
  28. 2020Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Networkcited by 24
  29. 2020SUMO: Unbiased Estimation of Log Marginal Probability for Latent Variable Modelscited by 14
  30. 2022How Tempering Fixes Data Augmentation in Bayesian Neural Networkscited by 13
  31. 2023Understanding the Detrimental Class-level Effects of Data Augmentationcited by 12
  32. 2024Posterior Uncertainty Quantification in Neural Networks using Data Augmentationcited by 9
  33. 2023Selective Mixup Helps with Distribution Shifts, But Not (Only) because of Mixupcited by 8
  34. 2023What Are Effective Labels for Augmented Data? Improving Calibration and Robustness with AutoLabelcited by 6
  35. 2023One-Line-of-Code Data Mollification Improves Optimization of Likelihood-based Generative Modelscited by 4
  36. 2019Data Augmentation for Bayesian Deep Learningcited by 2
  37. 2021Instance-based Label Smoothing For Better Calibrated Classification Networkscited by 2

Built on by

Cited by year

Citing papers per year: 2025: 2.

2 papers.

  1. 2025MaxSup: Overcoming Representation Collapse in Label Smoothingcited by 4
  2. 2025Optimizing Data Augmentation through Bayesian Model Selectioncited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2023DiffAug: A Diffuse-and-Denoise Augmentation for Training Robust Classifiers
  2. 2025Robustifying Diffusion-Denoised Smoothing Against Covariate Shift
  3. 2023Benchmarking Robustness to Text-Guided Corruptions
  4. 2024DiffuseMix: Label-Preserving Data Augmentation with Diffusion Models
  5. 2023Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels
  6. 2025Context-guided Responsible Data Augmentation with Diffusion Models
  7. 2025Robust Representation Consistency Model via Contrastive Denoising
  8. 2022Enhancing Diffusion-Based Image Synthesis with Robust Classifier Guidance