mlsanityData to 7 Oct 2026

Expanding the Role of Diffusion Models for Robust Classifier Training

arXiv 2602.19931 ↗First posted 23 Feb 2026cs.LG cs.CVcited by 0

On the map: Diffusion models → Robustness, privacy and adversarial aspects

Abstract (arXiv)

Incorporating diffusion-generated synthetic data into adversarial training (AT) has been shown to substantially improve the training of robust image classifiers. In this work, we extend the role of diffusion models beyond merely generating synthetic data, examining whether their internal representations, which encode meaningful features of the data, can provide additional benefits for robust classifier training. Through systematic experiments, we show that diffusion models offer representations that are both diverse and partially robust, and that explicitly incorporating diffusion representations as an auxiliary learning signal during AT consistently improves robustness across settings. Furthermore, our representation analysis indicates that incorporating diffusion models into AT encourages more disentangled features, while diffusion representations and diffusion-generated synthetic data play complementary roles in shaping representations. Experiments on CIFAR-10, CIFAR-100, and ImageNet validate these findings, demonstrating the effectiveness of jointly leveraging diffusion representations and synthetic data within AT.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 67 of its 76 references that are in the map. First posted on arXiv 23 Feb 2026.

Builds on

67 of 76 references.

Closest ancestors · 8

Same direction, newest first

  1. 2025Diffusion Models are Robust Pretrainers4 Nov
  2. 2024ADBM: Adversarial diffusion bridge model for reliable adversarial purification1 Aug
  3. 2024Your Diffusion Model is Secretly a Certifiably Robust Classifier4 Feb
  4. 2023Robust Classification via a Single Diffusion Model24 May
  5. 2023Better Diffusion Models Further Improve Adversarial Training9 Feb
  6. 2022Improving Adversarial Robustness by Contrastive Guided Diffusion Process18 Oct
  7. 2022(Certified!!) Adversarial Robustness for Free!21 Jun
  8. 2022Diffusion Models for Adversarial Purification16 May

Other references · 59

  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. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  6. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  7. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  8. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  9. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  10. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  11. 2022Scalable Diffusion Models with Transformerscited by 5,931
  12. 2013Intriguing properties of neural networkscited by 5,114
38 more
  1. 2022A ConvNet for the 2020scited by 3,545
  2. 2016Wide Residual Networkscited by 3,057
  3. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  4. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  5. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  6. 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,393
  7. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,359
  8. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  9. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  10. 2022Toy Models of Superpositioncited by 928
  11. 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformerscited by 707
  12. 2024Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Thinkcited by 666
  13. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  14. 2024Scaling and evaluating sparse autoencoderscited by 530
  15. 2020Overfitting in adversarially robust deep learningcited by 504
  16. 2023Symbolic Discovery of Optimization Algorithmscited by 423
  17. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 379
  18. 2025Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Modelscited by 376
  19. 2023Your Diffusion Model is Secretly a Zero-Shot Classifiercited by 301
  20. 2023Text-to-Image Diffusion Models are Zero-Shot Classifierscited by 140
  21. 2023Denoising Diffusion Autoencoders are Unified Self-supervised Learnerscited by 128
  22. 2023Diffusion Model as Representation Learnercited by 95
  23. 2024Deconstructing Denoising Diffusion Models for Self-Supervised Learningcited by 89
  24. 2025What matters for Representation Alignment: Global Information or Spatial Structure?cited by 76
  25. 2023Decoupled Kullback-Leibler Divergence Losscited by 58
  26. 2024Learning by Reconstruction Produces Uninformative Features For Perceptioncited by 54
  27. 2023Intriguing properties of generative classifierscited by 52
  28. 2022Understanding Robust Overfitting of Adversarial Training and Beyondcited by 50
  29. 2022Rank Diminishing in Deep Neural Networkscited by 47
  30. 2023Robust Principles: Architectural Design Principles for Adversarially Robust CNNscited by 46
  31. 2024TinyFusion: Diffusion Transformers Learned Shallowcited by 44
  32. 2024CleanDIFT: Diffusion Features without Noisecited by 42
  33. 2024Adversarial Robustness Limits via Scaling-Law and Human-Alignment Studiescited by 38
  34. 2023DreamTeacher: Pretraining Image Backbones with Deep Generative Modelscited by 37
  35. 2022On the Limitations of Stochastic Pre-processing Defensescited by 21
  36. 2025Unleashing Diffusion Transformers for Visual Correspondence by Modulating Massive Activationscited by 20
  37. 2023Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspectivecited by 18
  38. 2025Understanding Representation Dynamics of Diffusion Models via Low-Dimensional Modelingcited by 18

50 most cited of 59 All 59 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022Enhancing Diffusion-Based Image Synthesis with Robust Classifier Guidance
  2. 2024Efficient Image-to-Image Diffusion Classifier for Adversarial Robustness
  3. 2025What is Adversarial Training for Diffusion Models?
  4. 2022DensePure: Understanding Diffusion Models towards Adversarial Robustness
  5. 2025Demystifying Adversarial Robustness in Diffusion Models: Compression, Randomness, and Geometry
  6. 2024Adversarial Examples are Misaligned in Diffusion Model Manifolds
  7. 2025Diffusion or Non-Diffusion Adversarial Defenses: Rethinking the Relation between Classifier and Adversarial Purifier
  8. 2026Representation-Conditioned Diffusion Models for Guided Training Data Generation