mlsanityData to 7 Oct 2026

Detecting Adversarial Data by Probing Multiple Perturbations Using Expected Perturbation Score

arXiv 2305.16035 ↗First posted 25 May 2023cs.LG cs.CRcited by 20

Not on the map. Why?

Abstract (arXiv)

Adversarial detection aims to determine whether a given sample is an adversarial one based on the discrepancy between natural and adversarial distributions. Unfortunately, estimating or comparing two data distributions is extremely difficult, especially in high-dimension spaces. Recently, the gradient of log probability density (a.k.a., score) w.r.t. the sample is used as an alternative statistic to compute. However, we find that the score is sensitive in identifying adversarial samples due to insufficient information with one sample only. In this paper, we propose a new statistic called expected perturbation score (EPS), which is essentially the expected score of a sample after various perturbations. Specifically, to obtain adequate information regarding one sample, we perturb it by adding various noises to capture its multi-view observations. We theoretically prove that EPS is a proper statistic to compute the discrepancy between two samples under mild conditions. In practice, we can use a pre-trained diffusion model to estimate EPS for each sample. Last, we propose an EPS-based adversarial detection (EPS-AD) method, in which we develop EPS-based maximum mean discrepancy (MMD) as a metric to measure the discrepancy between the test sample and natural samples. We also prove that the EPS-based MMD between natural and adversarial samples is larger than that among natural samples. Extensive experiments show the superior adversarial detection performance of our EPS-AD.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Builds on 40 of its 62 references that are in the map. First posted on arXiv 25 May 2023.

Builds on

40 of 62 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  4. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  6. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  7. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  8. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  9. 2013Intriguing properties of neural networkscited by 5,114
  10. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  11. 2016Wide Residual Networkscited by 3,057
  12. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
28 more
  1. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  2. 2016Adversarial examples in the physical worldcited by 1,762
  3. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  4. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  5. 2015Learning Transferable Features with Deep Adaptation Networkscited by 1,323
  6. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  7. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  8. 2022Diffusion Models for Adversarial Purificationcited by 469
  9. 2022Efficient Test-Time Model Adaptation without Forgettingcited by 467
  10. 2019Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attackscited by 440
  11. 2023Towards Stable Test-Time Adaptation in Dynamic Wild Worldcited by 409
  12. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 280
  13. 2017Detecting Adversarial Samples from Artifactscited by 244
  14. 2021Improving Robustness using Generated Datacited by 233
  15. 2021A Variational Perspective on Diffusion-Based Generative Models and Score Matchingcited by 182
  16. 2017On the (Statistical) Detection of Adversarial Examplescited by 180
  17. 2020Perceptual Adversarial Robustness: Defense Against Unseen Threat Modelscited by 137
  18. 2021Adversarial purification with Score-based generative modelscited by 132
  19. 2020Learning Deep Kernels for Non-Parametric Two-Sample Testscited by 129
  20. 2022Improving Adversarial Transferability via Neuron Attribution-Based Attackscited by 92
  21. 2020Maximum Mean Discrepancy Test is Aware of Adversarial Attackscited by 44
  22. 2022Aligning Domain-specific Distribution and Classifier for Cross-domain Classification from Multiple Sourcescited by 42
  23. 2017A Linear-Time Kernel Goodness-of-Fit Testcited by 41
  24. 2021LiBRe: A Practical Bayesian Approach to Adversarial Detectioncited by 40
  25. 2019Impact of Adversarial Examples on Deep Learning Models for Biomedical Image Segmentationcited by 26
  26. 2020A General Framework For Detecting Anomalous Inputs to DNN Classifierscited by 24
  27. 2021Two Coupled Rejection Metrics Can Tell Adversarial Examples Apartcited by 14
  28. 2022Fast and Reliable Evaluation of Adversarial Robustness with Minimum-Margin Attackcited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 5; 2025: 9; 2026: 4.

20 papers.

Sort:

By area:

  1. 2024Detecting Machine-Generated Texts by Multi-Population Aware Optimization for Maximum Mean Discrepancycited by 20
  2. 2023Score Mismatching for Generative Modelingcited by 7
  3. 2024Improving Accuracy-robustness Trade-off via Pixel Reweighted Adversarial Trainingcited by 6
  4. 2025DUAL: Learning Diverse Kernels for Aggregated Two-sample and Independence Testingcited by 6
  5. 2025One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional Discrepancycited by 3
  6. 2025Sample-Specific Noise Injection For Diffusion-Based Adversarial Purificationcited by 2
  7. 2025Universal and Efficient Detection of Adversarial Data through Nonuniform Impact on Network Layerscited by 2
  8. 2024Detecting Adversarial Data using Perturbation Forgerycited by 1
  9. 2025Topological Signatures of Adversaries in Multimodal Alignmentscited by 1
  10. 2025Rectifying Adversarial Sample with Low Entropy Prior for Test-Time Defensecited by 1
  11. 2025Conformal Data Contamination Tests for In-distribution Data Acquisitioncited by 1
  12. 2025Generative Model Inversion Through the Lens of the Manifold Hypothesiscited by 1
8 more
  1. 2025Anchor-based Maximum Discrepancy for Relative Similarity Testingcited by 1
  2. 2023Parameter-Saving Adversarial Training: Reinforcing Multi-Perturbation Robustness via Hypernetworkscited by 0
  3. 2024Resilience and Security of Deep Neural Networks Against Intentional and Unintentional Perturbations: Survey and Research Challengescited by 0
  4. 2024Adversarial Detection with a Dynamically Stable Systemcited by 0
  5. 2026Advantage-based Temporal Attack in Reinforcement Learningcited by 0
  6. 2026S2O: Enhancing Adversarial Training with Second-Order Statistics of Weightscited by 0
  7. 2026Detecting Adversarial Data via Provable Adversarial Noise Amplificationcited by 0
  8. 2026USAD: Uncertainty-aware Statistical Adversarial Detectioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Sample-agnostic Adversarial Perturbation for Vision-Language Pre-training Models
  2. 2023Investigating the Adversarial Robustness of Density Estimation Using the Probability Flow ODE
  3. 2023Enhancing Adversarial Robustness via Score-Based Optimization
  4. 2025ScoreAdv: Score-based Targeted Generation of Natural Adversarial Examples via Diffusion Models
  5. 2023Assessing Robustness via Score-Based Adversarial Image Generation
  6. 2025Breaking Latent Prior Bias in Detectors for Generalizable AIGC Image Detection
  7. 2024Vulnerabilities in AI-generated Image Detection: The Challenge of Adversarial Attacks
  8. 2023Adversarial Counterfactual Visual Explanations