mlsanityData to 7 Oct 2026

Transcending Adversarial Perturbations: Manifold-Aided Adversarial Examples with Legitimate Semantics

arXiv 2402.03095 ↗First posted 5 Feb 2024cs.CV cs.CR cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Deep neural networks were significantly vulnerable to adversarial examples manipulated by malicious tiny perturbations. Although most conventional adversarial attacks ensured the visual imperceptibility between adversarial examples and corresponding raw images by minimizing their geometric distance, these constraints on geometric distance led to limited attack transferability, inferior visual quality, and human-imperceptible interpretability. In this paper, we proposed a supervised semantic-transformation generative model to generate adversarial examples with real and legitimate semantics, wherein an unrestricted adversarial manifold containing continuous semantic variations was constructed for the first time to realize a legitimate transition from non-adversarial examples to adversarial ones. Comprehensive experiments on MNIST and industrial defect datasets showed that our adversarial examples not only exhibited better visual quality but also achieved superior attack transferability and more effective explanations for model vulnerabilities, indicating their great potential as generic adversarial examples. The code and pre-trained models were available at https://github.com/shuaili1027/MAELS.git.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 31 of its 41 references that are in the map. First posted on arXiv 5 Feb 2024.

Builds on

31 of 41 references.

References · 31

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  4. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  5. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  6. 2016Adversarial Machine Learning at Scalecited by 934
  7. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  8. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  9. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  10. 2022Diffusion Models for Adversarial Purificationcited by 469
  11. 2020InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANscited by 393
  12. 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 328
19 more
  1. 2018Generating Adversarial Examples with Adversarial Networkscited by 256
  2. 2019The Secret Revealer: Generative Model-Inversion Attacks Against Deep Neural Networkscited by 241
  3. 2020A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promisescited by 230
  4. 2017Generating Natural Adversarial Examplescited by 182
  5. 2018Constructing Unrestricted Adversarial Examples with Generative Modelscited by 135
  6. 2019SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editingcited by 93
  7. 2019ColorFool: Semantic Adversarial Colorizationcited by 89
  8. 2018Semantic Adversarial Examplescited by 86
  9. 2020Use HiResCAM instead of Grad-CAM for faithful explanations of convolutional neural networkscited by 79
  10. 2018Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samplescited by 38
  11. 2022Query-Efficient Black-box Adversarial Attacks Guided by a Transfer-based Priorcited by 38
  12. 2017The Robust Manifold Defense: Adversarial Training using Generative Modelscited by 35
  13. 2021Robust Feature-Level Adversaries are Interpretability Toolscited by 27
  14. 2021Demiguise Attack: Crafting Invisible Semantic Adversarial Perturbations with Perceptual Similaritycited by 18
  15. 2020Breaking certified defenses: Semantic adversarial examples with spoofed robustness certificatescited by 17
  16. 2021Attack to Fool and Explain Deep Networkscited by 17
  17. 2022Discrete Point-wise Attack Is Not Enough: Generalized Manifold Adversarial Attack for Face Recognitioncited by 11
  18. 2021Semantic Perturbations with Normalizing Flows for Improved Generalizationcited by 6
  19. 2022On the Minimal Adversarial Perturbation for Deep Neural Networks with Provable Estimation Errorcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 1; 2026: 2.

4 papers.

  1. 2024Seeing is Deceiving: Exploitation of Visual Pathways in Multi-Modal Language Modelscited by 4
  2. 2025Beyond Vulnerabilities: A Survey of Adversarial Attacks as Both Threats and Defenses in Computer Vision Systemscited by 4
  3. 2026Beyond Waveform Robustness: Robust Feature-Vocoder Adversarial Attacks on Automatic Speech Recognitioncited by 0
  4. 2026Does Target Alignment Mean Target Recovery? An Evidence-Ladder Study of Adversarial Claims on Contrastive Encoderscited by 0