mlsanityData to 7 Oct 2026

Backpropagating Linearly Improves Transferability of Adversarial Examples

arXiv 2012.03528 ↗First posted 7 Dec 2020cs.LG cs.CR cs.CVcited by 79

Not on the map. Why?

Abstract (arXiv)

The vulnerability of deep neural networks (DNNs) to adversarial examples has drawn great attention from the community. In this paper, we study the transferability of such examples, which lays the foundation of many black-box attacks on DNNs. We revisit a not so new but definitely noteworthy hypothesis of Goodfellow et al.'s and disclose that the transferability can be enhanced by improving the linearity of DNNs in an appropriate manner. We introduce linear backpropagation (LinBP), a method that performs backpropagation in a more linear fashion using off-the-shelf attacks that exploit gradients. More specifically, it calculates forward as normal but backpropagates loss as if some nonlinear activations are not encountered in the forward pass. Experimental results demonstrate that this simple yet effective method obviously outperforms current state-of-the-arts in crafting transferable adversarial examples on CIFAR-10 and ImageNet, leading to more effective attacks on a variety of DNNs.

Key facts. Cited by 79 papers in the map (data through 7 Oct 2026). First posted on arXiv 7 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 15; 2022: 13; 2023: 23; 2024: 13; 2025: 6; 2026: 9.

79 papers.

Sort:

By area:

  1. 2021Advances in adversarial attacks and defenses in computer vision: A surveycited by 127
  2. 2021On Improving Adversarial Transferability of Vision Transformerscited by 70
  3. 2022Boosting the Transferability of Adversarial Attacks with Reverse Adversarial Perturbationcited by 67
  4. 2021On Generating Transferable Targeted Perturbationscited by 65
  5. 2023Structure Invariant Transformation for better Adversarial Transferabilitycited by 61
  6. 2021Adversarial Robustness Comparison of Vision Transformer and MLP-Mixer to CNNscited by 57
  7. 2023A Survey on Transferability of Adversarial Examples across Deep Neural Networkscited by 43
  8. 2022Towards Understanding and Boosting Adversarial Transferability from a Distribution Perspectivecited by 36
  9. 2022Transferable Adversarial Attack based on Integrated Gradientscited by 35
  10. 2021A Little Robustness Goes a Long Way: Leveraging Robust Features for Targeted Transfer Attackscited by 33
  11. 2023Making Substitute Models More Bayesian Can Enhance Transferability of Adversarial Examplescited by 32
  12. 2023Why Does Little Robustness Help? A Further Step Towards Understanding Adversarial Transferabilitycited by 32
38 more
  1. 2023Improving Adversarial Transferability via Intermediate-level Perturbation Decaycited by 29
  2. 2023Rethinking the Backward Propagation for Adversarial Transferabilitycited by 27
  3. 2022Investigating Top-k White-Box and Transferable Black-box Attackcited by 26
  4. 2021Adaptive Image Transformations for Transfer-based Adversarial Attackcited by 22
  5. 2023Attacks in Adversarial Machine Learning: A Systematic Survey from the Life-cycle Perspectivecited by 19
  6. 2023Logit Margin Matters: Improving Transferable Targeted Adversarial Attack by Logit Calibrationcited by 18
  7. 2024Improved Generation of Adversarial Examples Against Safety-aligned LLMscited by 17
  8. 2023Exploring Transferability of Multimodal Adversarial Samples for Vision-Language Pre-training Models with Contrastive Learningcited by 16
  9. 2021Training Meta-Surrogate Model for Transferable Adversarial Attackcited by 15
  10. 2021Adversarial Examples on Segmentation Models Can be Easy to Transfercited by 15
  11. 2022Demystifying the Adversarial Robustness of Random Transformation Defensescited by 14
  12. 2022Towards Good Practices in Evaluating Transfer Adversarial Attackscited by 14
  13. 2022Minimizing Maximum Model Discrepancy for Transferable Black-box Targeted Attackscited by 14
  14. 2021Learning to Learn Transferable Attackcited by 13
  15. 2023Reliable Evaluation of Adversarial Transferabilitycited by 12
  16. 2024DAT: Improving Adversarial Robustness via Generative Amplitude Mix-up in Frequency Domaincited by 12
  17. 2023StyLess: Boosting the Transferability of Adversarial Examplescited by 11
  18. 2022Squeeze Training for Adversarial Robustnesscited by 10
  19. 2023Improving Adversarial Transferability via Model Alignmentcited by 10
  20. 2022An Intermediate-level Attack Framework on The Basis of Linear Regressioncited by 9
  21. 2023Improving the Transferability of Adversarial Examples via Direction Tuningcited by 9
  22. 2024Practical Region-level Attack against Segment Anything Modelscited by 9
  23. 2021Graph-Based Similarity of Neural Network Representationscited by 7
  24. 2024Rethinking Impersonation and Dodging Attacks on Face Recognition Systemscited by 7
  25. 2023Boosting Adversarial Transferability via Fusing Logits of Top-1 Decomposed Featurecited by 6
  26. 2023SoK: Pitfalls in Evaluating Black-Box Attackscited by 6
  27. 2023Towards Evaluating Transfer-based Attacks Systematically, Practically, and Fairlycited by 6
  28. 2024Boosting Adversarial Transferability with Low-Cost Optimization via Maximin Expected Flatnesscited by 6
  29. 2023Attack-SAM: Towards Attacking Segment Anything Model With Adversarial Examplescited by 5
  30. 2023LRS: Enhancing Adversarial Transferability through Lipschitz Regularized Surrogatecited by 5
  31. 2024BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Modelscited by 5
  32. 2022Low-Mid Adversarial Perturbation against Unauthorized Face Recognition Systemcited by 4
  33. 2023Backpropagation Path Search On Adversarial Transferabilitycited by 4
  34. 2024Adversarial Attacks on Both Face Recognition and Face Anti-spoofing Modelscited by 4
  35. 2025Exploring Adversarial Transferability between Kolmogorov-arnold Networkscited by 4
  36. 2021Query Attack by Multi-Identity Surrogatescited by 3
  37. 2021Clustering Effect of (Linearized) Adversarial Robust Modelscited by 3
  38. 2023Improving Transferability of Adversarial Examples via Bayesian Attackscited by 3

50 most cited of 79 All 79 (JSON)