mlsanityData to 14 Oct 2026

Full spectrum Unlearnable Examples via Spectral Equalization

arXiv 2606.26719 ↗First posted 25 Jun 2026cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Unlearnable examples (UEs) protect training data by injecting imperceptible perturbations so that models fail to extract exploitable representations. In this paper, we reveal that existing UEs exhibit a critical failure once low-pass filtering is applied, indicating that the effective perturbation signals for unlearnability concentrate predominantly in high frequencies. Hence, we argue that reliable UEs should remain effective across the full spectrum. To this end, we propose Full-spectrum Unlearnable examples via Spectral Equalization (FUSE), which aims to generate spectrum-agnostic perturbations by equalizing the contributions from different bands and enforcing cross-band consistency. Specifically, FUSE adopts a Random Spectral Masking (RSM) strategy during generator training, which randomly removes a contiguous frequency band, forcing the remaining bands to maintain unlearnability. In addition, FUSE further integrates Cross-Band Guidance (CBG), which enforces mutual consistency between high- and low-frequency components, thereby further enhancing low-frequency unlearnability and regulating high-frequency perturbations to preserve the semantic fidelity of images. Extensive experiments across multiple datasets, architectures, and spectral filtering demonstrate the strong protection achieved by FUSE.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 41 of its 67 references that are in the map. First posted on arXiv 25 Jun 2026.

Builds on

41 of 67 references.

References · 41

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  5. 2016Densely Connected Convolutional Networkscited by 6,720
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  7. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  8. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
  9. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,628
  10. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
  11. 2021Unlearnable Examples: Making Personal Data Unexploitablecited by 193
  12. 2021Availability Attacks Create Shortcutscited by 62
29 more
  1. 2021Which Shortcut Cues Will DNNs Choose? A Study from the Parameter-Space Perspectivecited by 57
  2. 2023When Source-Free Domain Adaptation Meets Learning with Noisy Labelscited by 47
  3. 2022Autoregressive Perturbations for Data Poisoningcited by 46
  4. 2022Transferable Unlearnable Examplescited by 43
  5. 2022Unlearnable Clusters: Towards Label-agnostic Unlearnable Examplescited by 36
  6. 2022Robust Unlearnable Examples: Protecting Data Against Adversarial Learningcited by 35
  7. 2022On Learning Fairness and Accuracy on Multiple Subgroupscited by 32
  8. 2022One-Pixel Shortcut: on the Learning Preference of Deep Neural Networkscited by 29
  9. 2023What Can We Learn from Unlearnable Datasets?cited by 21
  10. 2024Multimodal Unlearnable Examples: Protecting Data against Multimodal Contrastive Learningcited by 21
  11. 2024Transferable Adversarial Attacks on SAM and Its Downstream Modelscited by 21
  12. 2025On the Benefits of Attribute-Driven Graph Domain Adaptationcited by 19
  13. 2024Purify Unlearnable Examples via Rate-Constrained Variational Autoencoderscited by 18
  14. 2024Provably Unlearnable Data Examplescited by 18
  15. 2022Structure-Preserving Graph Representation Learningcited by 17
  16. 2024EAGLE: Elevating Geometric Reasoning through LLM-empowered Visual Instruction Tuningcited by 16
  17. 2024Unlearnable 3D Point Clouds: Class-wise Transformation Is All You Needcited by 16
  18. 2023Detection and Defense of Unlearnable Examplescited by 13
  19. 2024Generalizing across Temporal Domains with Koopman Operatorscited by 13
  20. 2024DAT: Improving Adversarial Robustness via Generative Amplitude Mix-up in Frequency Domaincited by 12
  21. 2025ZETA: Leveraging Z-order Curves for Efficient Top-k Attentioncited by 9
  22. 2025Homophily Enhanced Graph Domain Adaptationcited by 9
  23. 2023Progressive Text-to-Image Diffusion with Soft Latent Directioncited by 6
  24. 2024Learning from Noisy Labels via Conditional Distributionally Robust Optimizationcited by 6
  25. 2024Intersectional Unfairness Discoverycited by 5
  26. 2026When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretrainingcited by 5
  27. 2026Graph Domain Adaptation via Homophily-Agnostic Reconstructing Structurecited by 3
  28. 2026Dual-branch Robust Unlearnable Examplescited by 2
  29. 2026Addressing both variable selection and misclassified responses with parametric and semiparametric methodscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Unlearnable Faces: Privacy Protection Surviving Extraction Pipelinecited by 0
  2. 2026MUGEN: Generating Unlearnable Graph Examples for Multiple Learning Taskscited by 0
  3. 2026ARM: Attention with Routed-Memory for Learnable Sparse Controlcited by 0