mlsanityData to 7 Oct 2026

A Survey On Universal Adversarial Attack

arXiv 2103.01498 ↗First posted 2 Mar 2021cs.LG cs.CVcited by 59

Not on the map. Why?

Abstract (arXiv)

The intriguing phenomenon of adversarial examples has attracted significant attention in machine learning and what might be more surprising to the community is the existence of universal adversarial perturbations (UAPs), i.e. a single perturbation to fool the target DNN for most images. With the focus on UAP against deep classifiers, this survey summarizes the recent progress on universal adversarial attacks, discussing the challenges from both the attack and defense sides, as well as the reason for the existence of UAP. We aim to extend this work as a dynamic survey that will regularly update its content to follow new works regarding UAP or universal attack in a wide range of domains, such as image, audio, video, text, etc. Relevant updates will be discussed at: https://bit.ly/2SbQlLG. We welcome authors of future works in this field to contact us for including your new finding.

Key facts. Cited by 59 papers in the map (data through 7 Oct 2026). Builds on 47 of its 61 references that are in the map. First posted on arXiv 2 Mar 2021.

Builds on

47 of 61 references.

References · 47

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  5. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  6. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  7. 2016Adversarial Machine Learning at Scalecited by 934
  8. 2016Universal adversarial perturbationscited by 774
  9. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  10. 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 709
  11. 2018Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Surveycited by 459
  12. 2018Adversarially Robust Generalization Requires More Datacited by 318
35 more
  1. 2019You Only Propagate Once: Accelerating Adversarial Training via Maximal Principlecited by 181
  2. 2017Generative Adversarial Perturbationscited by 153
  3. 2018Are adversarial examples inevitable?cited by 113
  4. 2017Universal Adversarial Perturbations Against Semantic Image Segmentationcited by 92
  5. 2019Cross-Domain Transferability of Adversarial Perturbationscited by 92
  6. 2018Generalizable Data-free Objective for Crafting Universal Adversarial Perturbationscited by 88
  7. 2016Robustness of classifiers: from adversarial to random noisecited by 87
  8. 2018Universal Adversarial Trainingcited by 85
  9. 2017Fast Feature Fool: A data independent approach to universal adversarial perturbationscited by 65
  10. 2020Understanding Adversarial Examples from the Mutual Influence of Images and Perturbationscited by 59
  11. 2017NAG: Network for Adversary Generationcited by 51
  12. 2020Universal Adversarial Attacks with Natural Triggers for Text Classificationcited by 50
  13. 2017Art of singular vectors and universal adversarial perturbationscited by 49
  14. 2017Defense against Universal Adversarial Perturbationscited by 48
  15. 2019Universal Adversarial Perturbations for Speech Recognition Systemscited by 41
  16. 2018Adversarial Perturbations Against Real-Time Video Classification Systemscited by 32
  17. 2018Ask, Acquire, and Attack: Data-free UAP Generation using Class Impressionscited by 31
  18. 2018Universal Perturbation Attack Against Image Retrievalcited by 31
  19. 2020Enabling Fast and Universal Audio Adversarial Attack Using Generative Modelcited by 31
  20. 2020CD-UAP: Class Discriminative Universal Adversarial Perturbationcited by 27
  21. 2020Real-time, Universal, and Robust Adversarial Attacks Against Speaker Recognition Systemscited by 26
  22. 2018Defending Against Universal Perturbations With Shared Adversarial Trainingcited by 25
  23. 2020Double Targeted Universal Adversarial Perturbationscited by 23
  24. 2019Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defensescited by 22
  25. 2021Universal Adversarial Perturbations Through the Lens of Deep Steganography: Towards A Fourier Perspectivecited by 18
  26. 2019Appending Adversarial Frames for Universal Video Attackcited by 17
  27. 2017Robustness of classifiers to universal perturbations: a geometric perspectivecited by 14
  28. 2020Universal Adversarial Perturbations Generative Network for Speaker Recognitioncited by 13
  29. 2020Transferable Universal Adversarial Perturbations Using Generative Modelscited by 10
  30. 2018Playing the Game of Universal Adversarial Perturbationscited by 9
  31. 2019Universal Adversarial Perturbation for Text Classificationcited by 8
  32. 2020Decision-based Universal Adversarial Attackcited by 6
  33. 2019A Method for Computing Class-wise Universal Adversarial Perturbationscited by 5
  34. 2019Universal adversarial examples in speech command classificationcited by 4
  35. 2019Fast-UAP: An Algorithm for Speeding up Universal Adversarial Perturbation Generation with Orientation of Perturbation Vectorscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 18; 2023: 11; 2024: 12; 2025: 8; 2026: 3.

59 papers.

Sort:

By area:

  1. 2021Ethical and social risks of harm from Language Modelscited by 957
  2. 2023Large Language Model Alignment: A Surveycited by 241
  3. 2022Physical Adversarial Attack meets Computer Vision: A Decade Surveycited by 66
  4. 2021Adversarial Robustness Comparison of Vision Transformer and MLP-Mixer to CNNscited by 57
  5. 2023Strategic Preys Make Acute Predators: Enhancing Camouflaged Object Detectors by Generating Camouflaged Objectscited by 56
  6. 2024Test-Time Backdoor Attacks on Multimodal Large Language Modelscited by 35
  7. 2022Phantom Sponges: Exploiting Non-Maximum Suppression to Attack Deep Object Detectorscited by 28
  8. 2022Gradient Methods Provably Converge to Non-Robust Networkscited by 27
  9. 2022Investigating Top-k White-Box and Transferable Black-box Attackcited by 26
  10. 2023The Double-Edged Sword of Implicit Bias: Generalization vs. Robustness in ReLU Networkscited by 23
  11. 2022A Survey of Adversarial Defences and Robustness in NLPcited by 22
  12. 2026Hidden in Memory: Sleeper Memory Poisoning in LLM Agentscited by 22
38 more
  1. 2022Adversarial Patch Attacks and Defences in Vision-Based Tasks: A Surveycited by 17
  2. 2022AI Ethics Issues in Real World: Evidence from AI Incident Databasecited by 14
  3. 2023Mutual-modality Adversarial Attack with Semantic Perturbationcited by 14
  4. 2021Universal Adversarial Training with Class-Wise Perturbationscited by 13
  5. 2021Black-box adversarial attacks using Evolution Strategiescited by 11
  6. 2024Ungeneralizable Examplescited by 9
  7. 2024Texture Re-scalable Universal Adversarial Perturbationcited by 9
  8. 2021Deep Image Destruction: Vulnerability of Deep Image-to-Image Models against Adversarial Attackscited by 8
  9. 2025PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomaliescited by 8
  10. 2021Semantically Stealthy Adversarial Attacks against Segmentation Modelscited by 7
  11. 2024GE-AdvGAN: Improving the transferability of adversarial samples by gradient editing-based adversarial generative modelcited by 7
  12. 2024Image-to-Text Logic Jailbreak: Your Imagination can Help You Do Anythingcited by 7
  13. 2022Consistent Attack: Universal Adversarial Perturbation on Embodied Vision Navigationcited by 6
  14. 2022Physical Passive Patch Adversarial Attacks on Visual Odometry Systemscited by 6
  15. 2023Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systemscited by 6
  16. 2025FairMedQA: Benchmarking Bias in Large Language Models for Medical Question Answeringcited by 6
  17. 2022On the Robustness of Explanations of Deep Neural Network Models: A Surveycited by 5
  18. 2021MINIMAL: Mining Models for Data Free Universal Adversarial Triggerscited by 4
  19. 2022On the uncertainty principle of neural networkscited by 4
  20. 2023Comparative Evaluation of Recent Universal Adversarial Perturbations in Image Classificationcited by 4
  21. 2024Quantum-Inspired Analysis of Neural Network Vulnerabilities: The Role of Conjugate Variables in System Attackscited by 4
  22. 2025XOXO: Stealthy Cross-Origin Context Poisoning Attacks against AI Coding Assistantscited by 4
  23. 2023AdaptGuard: Defending Against Universal Attacks for Model Adaptationcited by 3
  24. 2024Exploring adversarial robustness of JPEG AI: methodology, comparison and new methodscited by 3
  25. 2022GUARD: Graph Universal Adversarial Defensecited by 2
  26. 2023On the Universal Adversarial Perturbations for Efficient Data-free Adversarial Detectioncited by 2
  27. 2024Edge-Only Universal Adversarial Attacks in Distributed Learningcited by 2
  28. 2022A Simple Structure For Building A Robust Modelcited by 1
  29. 2022Adversarial Noises Are Linearly Separable for (Nearly) Random Neural Networkscited by 1
  30. 2022Unifying Gradients to Improve Real-world Robustness for Deep Networkscited by 1
  31. 2022Chaos Theory and Adversarial Robustnesscited by 1
  32. 2023Adversarial AI in Insurance: Pervasiveness and Resiliencecited by 1
  33. 2023Magmaw: Modality-Agnostic Adversarial Attacks on Machine Learning-Based Wireless Communication Systemscited by 1
  34. 2024You Only Need Half: Boosting Data Augmentation by Using Partial Contentcited by 1
  35. 2025A Survey of Adversarial Defenses in Vision-based Systems: Categorization, Methods and Challengescited by 1
  36. 2025NIC-RobustBench: A Comprehensive Open-Source Toolkit for Neural Image Compression and Robustness Analysiscited by 1
  37. 2022Learning from Attacks: Attacking Variational Autoencoder for Improving Image Classificationcited by 0
  38. 2022Adversarial Detection by Approximation of Ensemble Boundarycited by 0

50 most cited of 59 All 59 (JSON)