mlsanityData to 7 Oct 2026

Improving Adversarial Robustness by Contrastive Guided Diffusion Process

arXiv 2210.09643 ↗First posted 18 Oct 2022cs.LG cs.CVcited by 8

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

Abstract (arXiv)

Synthetic data generation has become an emerging tool to help improve the adversarial robustness in classification tasks since robust learning requires a significantly larger amount of training samples compared with standard classification tasks. Among various deep generative models, the diffusion model has been shown to produce high-quality synthetic images and has achieved good performance in improving the adversarial robustness. However, diffusion-type methods are typically slow in data generation as compared with other generative models. Although different acceleration techniques have been proposed recently, it is also of great importance to study how to improve the sample efficiency of generated data for the downstream task. In this paper, we first analyze the optimality condition of synthetic distribution for achieving non-trivial robust accuracy. We show that enhancing the distinguishability among the generated data is critical for improving adversarial robustness. Thus, we propose the Contrastive-Guided Diffusion Process (Contrastive-DP), which adopts the contrastive loss to guide the diffusion model in data generation. We verify our theoretical results using simulations and demonstrate the good performance of Contrastive-DP on image datasets.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 47 of its 64 references that are in the map. First posted on arXiv 18 Oct 2022.

Builds on

47 of 64 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022Diffusion Models for Adversarial Purification16 May

Other references · 46

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2020Denoising Diffusion Implicit Modelscited by 9,655
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  6. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  7. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  8. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  9. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  10. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  11. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  12. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
34 more
  1. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  2. 2016Gaussian Error Linear Units (GELUs)cited by 3,431
  3. 2016Wide Residual Networkscited by 3,057
  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. 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,194
  7. 2022Diffusion Models: A Comprehensive Survey of Methods and Applicationscited by 1,107
  8. 2020Improved Techniques for Training Score-Based Generative Modelscited by 1,036
  9. 2018Robustness May Be at Odds with Accuracycited by 873
  10. 2020Contrastive Learning with Hard Negative Samplescited by 619
  11. 2016Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithmcited by 498
  12. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  13. 2019Sliced Score Matching: A Scalable Approach to Density and Score Estimationcited by 381
  14. 2020Debiased Contrastive Learningcited by 354
  15. 2022Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Modelscited by 331
  16. 2018Adversarially Robust Generalization Requires More Datacited by 318
  17. 2020Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examplescited by 275
  18. 2021Improving Robustness using Generated Datacited by 233
  19. 2022Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Qualitycited by 181
  20. 2020Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Datacited by 179
  21. 2017Stein Variational Gradient Descent as Gradient Flowcited by 149
  22. 2020Adversarial Self-Supervised Contrastive Learningcited by 138
  23. 2021Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?cited by 113
  24. 2018Scalable End-to-End Autonomous Vehicle Testing via Rare-event Simulationcited by 76
  25. 2021When Does Contrastive Learning Preserve Adversarial Robustness from Pretraining to Finetuning?cited by 76
  26. 2019Bayesian Generative Active Deep Learningcited by 57
  27. 2018Learning To Simulatecited by 43
  28. 2019Sample-Efficient Neural Architecture Search by Learning Action Spacecited by 23
  29. 2020Improving Adversarial Robustness via Unlabeled Out-of-Domain Datacited by 23
  30. 2022Decoupled Adversarial Contrastive Learning for Self-supervised Adversarial Robustnesscited by 20
  31. 2021Conditional Synthetic Data Generation for Robust Machine Learning Applications with Limited Pandemic Datacited by 18
  32. 2022PointDP: Diffusion-driven Purification against Adversarial Attacks on 3D Point Cloud Recognitioncited by 16
  33. 2022Accelerating Score-based Generative Models for High-Resolution Image Synthesiscited by 10
  34. 2020LADA: Look-Ahead Data Acquisition via Augmentation for Active Learningcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 3; 2025: 3; 2026: 1.

8 papers.

  1. 2024A Survey on Statistical Theory of Deep Learning: Approximation, Training Dynamics, and Generative Modelscited by 20
  2. 2023Utility Theory of Synthetic Data Generationcited by 8
  3. 2024DiffBreak: Is Diffusion-Based Purification Robust?cited by 6
  4. 2024Diffusion Twigs with Loop Guidance for Conditional Graph Generationcited by 4
  5. 2025Demystifying Adversarial Robustness in Diffusion Models: Compression, Randomness, and Geometrycited by 1
  6. 2025Contrastive Conditional-Unconditional Alignment for Long-tailed Diffusion Modelcited by 1
  7. 2025Calibrating Uncertainty for Zero-Shot Adversarial CLIPcited by 1
  8. 2026Expanding the Role of Diffusion Models for Robust Classifier Trainingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Improved Diffusion-based Generative Model with Better Adversarial Robustness
  2. 2022Enhancing Diffusion-Based Image Synthesis with Robust Classifier Guidance
  3. 2025CAT: Contrastive Adversarial Training for Evaluating the Robustness of Protective Perturbations in Latent Diffusion Models
  4. 2023Robust Classification via a Single Diffusion Model
  5. 2023A High-Quality Robust Diffusion Framework for Corrupted Dataset
  6. 2024Adversarial Robustification via Text-to-Image Diffusion Models
  7. 2023On the Robustness of Latent Diffusion Models
  8. 2025What is Adversarial Training for Diffusion Models?