mlsanityData to 14 Oct 2026

Rethinking Target Label Conditioning in Adversarial Attacks: A 2D Tensor-Guided Generative Approach

arXiv 2504.14137 ↗First posted 19 Apr 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Compared to single-target adversarial attacks, multi-target attacks have garnered significant attention due to their ability to generate adversarial images for multiple target classes simultaneously. However, existing generative approaches for multi-target attacks primarily encode target labels into one-dimensional tensors, leading to a loss of fine-grained visual information and overfitting to model-specific features during noise generation. To address this gap, we first identify and validate that the semantic feature quality and quantity are critical factors affecting the transferability of targeted attacks: 1) Feature quality refers to the structural and detailed completeness of the implanted target features, as deficiencies may result in the loss of key discriminative information; 2) Feature quantity refers to the spatial sufficiency of the implanted target features, as inadequacy limits the victim model's attention to this feature. Based on these findings, we propose the 2D Tensor-Guided Adversarial Fusion (TGAF) framework, which leverages the powerful generative capabilities of diffusion models to encode target labels into two-dimensional semantic tensors for guiding adversarial noise generation. Additionally, we design a novel masking strategy tailored for the training process, ensuring that parts of the generated noise retain complete semantic information about the target class. Extensive experiments demonstrate that TGAF consistently surpasses state-of-the-art methods across various settings.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 55 of its 69 references that are in the map. First posted on arXiv 19 Apr 2025.

Builds on

55 of 69 references.

References · 55

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  6. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  7. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  8. 2020Denoising Diffusion Implicit Modelscited by 9,781
  9. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  10. 2016Densely Connected Convolutional Networkscited by 6,720
  11. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  12. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
38 more
  1. 2017Squeeze-and-Excitation Networkscited by 4,789
  2. 2014Going Deeper with Convolutionscited by 4,241
  3. 2016Identity Mappings in Deep Residual Networkscited by 1,965
  4. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,512
  5. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
  6. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 901
  7. 2017Ensemble Adversarial Training: Attacks and Defensescited by 840
  8. 2016Universal adversarial perturbationscited by 784
  9. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 599
  10. 2017Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networkscited by 486
  11. 2019Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attackscited by 442
  12. 2024The Platonic Representation Hypothesiscited by 382
  13. 2019Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attackscited by 355
  14. 2017Defense against Adversarial Attacks Using High-Level Representation Guided Denoisercited by 329
  15. 2021Enhancing the Transferability of Adversarial Attacks through Variance Tuningcited by 244
  16. 2016A study of the effect of JPG compression on adversarial imagescited by 197
  17. 2020A Self-supervised Approach for Adversarial Robustnesscited by 180
  18. 2017Generative Adversarial Perturbationscited by 153
  19. 2019Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systemscited by 153
  20. 2021Visformer: The Vision-friendly Transformercited by 137
  21. 2023Diffusion Models for Imperceptible and Transferable Adversarial Attackcited by 94
  22. 2019Cross-Domain Transferability of Adversarial Perturbationscited by 92
  23. 2020On Success and Simplicity: A Second Look at Transferable Targeted Attackscited by 89
  24. 2019PhysGAN: Generating Physical-World-Resilient Adversarial Examples for Autonomous Drivingcited by 72
  25. 2019advertorch v0.1: An Adversarial Robustness Toolbox based on PyTorchcited by 69
  26. 2021On Generating Transferable Targeted Perturbationscited by 65
  27. 2023Boosting Adversarial Transferability by Block Shuffle and Rotationcited by 64
  28. 2020Understanding Adversarial Examples from the Mutual Influence of Images and Perturbationscited by 59
  29. 2022Improving the Transferability of Targeted Adversarial Examples through Object-Based Diverse Inputcited by 50
  30. 2023AdvDiff: Generating Unrestricted Adversarial Examples using Diffusion Modelscited by 46
  31. 2020Yet Another Intermediate-Level Attackcited by 35
  32. 2021Boosting Transferability of Targeted Adversarial Examples via Hierarchical Generative Networkscited by 32
  33. 2022Enhancing the Self-Universality for Transferable Targeted Attackscited by 30
  34. 2024Learning to Transform Dynamically for Better Adversarial Transferabilitycited by 23
  35. 2021Feature Space Targeted Attacks by Statistic Alignmentcited by 21
  36. 2024CLIP-Guided Generative Networks for Transferable Targeted Adversarial Attackscited by 19
  37. 2023Logit Margin Matters: Improving Transferable Targeted Adversarial Attack by Logit Calibrationcited by 18
  38. 2023Introducing Competition to Boost the Transferability of Targeted Adversarial Examples through Clean Feature Mixupcited by 17

50 most cited of 55 All 55 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Semantically Guided Adversarial Testing of Vision Models Using Language Models
  2. 2023Semantic Adversarial Attacks via Diffusion Models
  3. 2026A Survey on Adversarial Attacks and Defenses for Diffusion Models Across Multiple Modalities
  4. 2024Efficient Generation of Targeted and Transferable Adversarial Examples for Vision-Language Models Via Diffusion Models
  5. 2025V-Attack: Targeting Disentangled Value Features for Controllable Adversarial Attacks on LVLMs
  6. 2025A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1
  7. 2025AdvAD: Exploring Non-Parametric Diffusion for Imperceptible Adversarial Attacks
  8. 2024AdvI2I: Adversarial Image Attack on Image-to-Image Diffusion models