mlsanityData to 7 Oct 2026

Fast Preemptive Robustification: High-Frequency Response Anti-Aligns Shared Vulnerability

arXiv 2603.25244 ↗First posted 26 Mar 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Adversarial attacks can readily compromise deep neural networks (DNNs). In particular, transferable attacks (TAs) exploit the shared vulnerabilities among DNNs, enabling perturbations crafted on surrogates to transfer to unseen models. Training-time and post-attack defenses have been extensively studied for combating TAs. Orthogonal to these approaches, preemptive robustification (PR) has emerged as a pre-attack defense that enhances the robustness of benign samples by superimposing protective variations before attacks. Despite its promise, PR remains underexplored and faces several important limitations. First, dependence on well-trained surrogate classifiers limits applicability, as surrogates are task-specific and may even be unavailable in some practical settings. Second, the required iterative optimization or dedicated PR generator training incurs substantial costs. Third, the generated variations are opaque to humans. To address these, we seek an efficient PR that is surrogate-free, optimization-free, training-free, and human-interpretable. Intriguingly, we discover a numerical correlation between the shared vulnerabilities of DNNs and Laplacian responses, with their cosine similarity being significantly negative. This indicates that negated high-frequency response constitutes an important component of shared vulnerabilities. Consequently, strengthening Laplacian responses counteracts this component, improving resistance to TAs. Building upon this insight, we propose Fast Preemptive Robustification (FPR), which performs Laplacian sharpening via a single channel-wise convolution with a 3×3 kernel. FPR is simple yet effective, as demonstrated by extensive experiments. Specifically, FPR reduces the attack success rate (ASR) of untargeted TAs by 12.7% and that of targeted TAs from 10.7% to 4.1%. Code will be released publicly.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 40 of its 65 references that are in the map. First posted on arXiv 26 Mar 2026.

Builds on

40 of 65 references.

References · 40

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2014Going Deeper with Convolutionscited by 4,231
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  8. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 530
  9. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  10. 2019High Frequency Component Helps Explain the Generalization of Convolutional Neural Networkscited by 310
  11. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  12. 2023Rethinking Model Ensemble in Transfer-based Adversarial Attackscited by 88
28 more
  1. 2023Revisiting Adversarial Training for ImageNet: Architectures, Training and Generalization across Threat Modelscited by 79
  2. 2019Image Super-Resolution as a Defense Against Adversarial Attackscited by 64
  3. 2023Boosting Adversarial Transferability by Block Shuffle and Rotationcited by 64
  4. 2023Robust Principles: Architectural Design Principles for Adversarially Robust CNNscited by 46
  5. 2023An Adaptive Model Ensemble Adversarial Attack for Boosting Adversarial Transferabilitycited by 42
  6. 2022Enhancing the Self-Universality for Transferable Targeted Attackscited by 30
  7. 2022Boosting the Transferability of Adversarial Attacks with Global Momentum Initializationcited by 29
  8. 2024MeanSparse: Post-Training Robustness Enhancement Through Mean-Centered Feature Sparsificationcited by 20
  9. 2023Logit Margin Matters: Improving Transferable Targeted Adversarial Attack by Logit Calibrationcited by 18
  10. 2023Introducing Competition to Boost the Transferability of Targeted Adversarial Examples through Clean Feature Mixupcited by 17
  11. 2024MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifierscited by 16
  12. 2022Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learningcited by 15
  13. 2024Instant Adversarial Purification with Adversarial Consistency Distillationcited by 14
  14. 2024Reducing Texture Bias of Deep Neural Networks via Edge Enhancing Diffusioncited by 11
  15. 2023Improving Adversarial Transferability via Model Alignmentcited by 10
  16. 2020Local Convolutions Cause an Implicit Bias towards High Frequency Adversarial Examplescited by 9
  17. 2024Enhancing targeted transferability via feature space fine-tuningcited by 7
  18. 2025Diffusion-based Adversarial Purification from the Perspective of the Frequency Domaincited by 7
  19. 2024Improving the Transferability of Adversarial Examples by Feature Augmentationcited by 5
  20. 2021Preemptive Image Robustification for Protecting Users against Man-in-the-Middle Adversarial Attackscited by 4
  21. 2025AIM: Additional Image Guided Generation of Transferable Adversarial Attackscited by 4
  22. 2025Harnessing the Computation Redundancy in ViTs to Boost Adversarial Transferabilitycited by 4
  23. 2025AntiDote: Bi-level Adversarial Training for Tamper-Resistant LLMscited by 4
  24. 2024Data-Driven Lipschitz Continuity: A Cost-Effective Approach to Improve Adversarial Robustnesscited by 3
  25. 2025Test-time augmentation improves efficiency in conformal predictioncited by 3
  26. 2025Boosting Adversarial Transferability via Residual Perturbation Attackcited by 1
  27. 2025NAT: Learning to Attack Neurons for Enhanced Adversarial Transferabilitycited by 1
  28. 2026OTI: A Model-free and Visually Interpretable Measure of Image Attackabilitycited by 1

Built on by

Not cited yet.