mlsanityData to 14 Oct 2026

Availability Attacks Create Shortcuts

arXiv 2111.00898 ↗First posted 1 Nov 2021cs.LG cs.CRcited by 62

Not on the map. Why?

Abstract (arXiv)

Availability attacks, which poison the training data with imperceptible perturbations, can make the data not exploitable by machine learning algorithms so as to prevent unauthorized use of data. In this work, we investigate why these perturbations work in principle. We are the first to unveil an important population property of the perturbations of these attacks: they are almost linearly separable when assigned with the target labels of the corresponding samples, which hence can work as shortcuts for the learning objective. We further verify that linear separability is indeed the workhorse for availability attacks. We synthesize linearly-separable perturbations as attacks and show that they are as powerful as the deliberately crafted attacks. Moreover, such synthetic perturbations are much easier to generate. For example, previous attacks need dozens of hours to generate perturbations for ImageNet while our algorithm only needs several seconds. Our finding also suggests that the shortcut learning is more widely present than previously believed as deep models would rely on shortcuts even if they are of an imperceptible scale and mixed together with the normal features. Our source code is published at urlhttps://github.com/dayu11/Availability-Attacks-Create-Shortcuts.

Key facts. Cited by 62 papers in the map (data through 14 Oct 2026). Builds on 37 of its 67 references that are in the map. First posted on arXiv 1 Nov 2021.

Builds on

37 of 67 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  5. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  6. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  7. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  8. 2016Densely Connected Convolutional Networkscited by 6,720
  9. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  10. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  11. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  12. 2018Neural Tangent Kernel: Convergence and Generalization in Neural Networkscited by 2,621
25 more
  1. 2020Extracting Training Data from Large Language Modelscited by 2,189
  2. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
  3. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,628
  4. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,512
  5. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  6. 2018Recognition in Terra Incognitacited by 658
  7. 2018Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networkscited by 519
  8. 2014Learning Face Representation from Scratchcited by 453
  9. 2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 414
  10. 2019Hidden Trigger Backdoor Attackscited by 371
  11. 2020The Pitfalls of Simplicity Bias in Neural Networkscited by 351
  12. 2020Input-Aware Dynamic Backdoor Attackcited by 313
  13. 2016Understanding Deep Neural Networks with Rectified Linear Unitscited by 257
  14. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  15. 2019Probing Neural Network Comprehension of Natural Language Argumentscited by 214
  16. 2021Unlearnable Examples: Making Personal Data Unexploitablecited by 193
  17. 2020Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defensescited by 180
  18. 2020Witches' Brew: Industrial Scale Data Poisoning via Gradient Matchingcited by 163
  19. 2020What shapes feature representations? Exploring datasets, architectures, and trainingcited by 126
  20. 2021Adversarial Examples Make Strong Poisonscited by 119
  21. 2020Just How Toxic is Data Poisoning? A Unified Benchmark for Backdoor and Data Poisoning Attackscited by 116
  22. 2020An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networkscited by 97
  23. 2021LowKey: Leveraging Adversarial Attacks to Protect Social Media Users from Facial Recognitioncited by 94
  24. 2019Learning to Confuse: Generating Training Time Adversarial Data with Auto-Encodercited by 70
  25. 2021Preventing Unauthorized Use of Proprietary Data: Poisoning for Secure Dataset Releasecited by 35

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 3; 2023: 18; 2024: 23; 2025: 8; 2026: 10.

62 papers.

Sort:

By area:

  1. 2023MetaCloak: Preventing Unauthorized Subject-driven Text-to-image Diffusion-based Synthesis via Meta-learningcited by 44
  2. 2023Image Shortcut Squeezing: Countering Perturbative Availability Poisons with Compressioncited by 39
  3. 2023Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitationcited by 37
  4. 2022Unlearnable Clusters: Towards Label-agnostic Unlearnable Examplescited by 36
  5. 2022Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learningcited by 34
  6. 2023Learning the Unlearnable: Adversarial Augmentations Suppress Unlearnable Example Attackscited by 30
  7. 2023CUDA: Convolution-based Unlearnable Datasetscited by 26
  8. 2022Indiscriminate Data Poisoning Attacks on Neural Networkscited by 25
  9. 2023Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning Attackscited by 23
  10. 2023What Can We Learn from Unlearnable Datasets?cited by 21
  11. 2023The Devil's Advocate: Shattering the Illusion of Unexploitable Data using Diffusion Modelscited by 19
  12. 2023A Survey of Robustness and Safety of 2D and 3D Deep Learning Models Against Adversarial Attackscited by 19
38 more
  1. 2024Purify Unlearnable Examples via Rate-Constrained Variational Autoencoderscited by 18
  2. 2024Provably Unlearnable Data Examplescited by 18
  3. 2023Unlearnable Examples Give a False Sense of Security: Piercing through Unexploitable Data with Learnable Examplescited by 17
  4. 2024UnSeg: One Universal Unlearnable Example Generator is Enough against All Image Segmentationcited by 14
  5. 2024Semantic Deep Hiding for Robust Unlearnable Examplescited by 13
  6. 2024Efficient Availability Attacks against Supervised and Contrastive Learning Simultaneouslycited by 12
  7. 2025MTL-UE: Learning to Learn Nothing for Multi-Task Learningcited by 11
  8. 2023Sharpness-Aware Data Poisoning Attackcited by 10
  9. 2024Disguised Copyright Infringement of Latent Diffusion Modelscited by 10
  10. 2024ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purificationcited by 9
  11. 2024Unlearnable Examples Detection via Iterative Filteringcited by 9
  12. 2025Temporal Unlearnable Examples: Preventing Personal Video Data from Unauthorized Exploitation by Object Trackingcited by 9
  13. 2024Indiscriminate Data Poisoning Attacks on Pre-trained Feature Extractorscited by 8
  14. 2024Medical Unlearnable Examples: Securing Medical Data from Unauthorized Training via Sparsity-Aware Local Maskingcited by 8
  15. 2023APBench: A Unified Benchmark for Availability Poisoning Attacks and Defensescited by 7
  16. 2024Toward Availability Attacks in 3D Point Cloudscited by 7
  17. 2023Segue: Side-information Guided Generative Unlearnable Examples for Facial Privacy Protection in Real Worldcited by 6
  18. 2024Adversarial Training: A Surveycited by 6
  19. 2025SafeSpeech: Robust and Universal Voice Protection Against Malicious Speech Synthesiscited by 6
  20. 2023Transferable Availability Poisoning Attackscited by 5
  21. 2026When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretrainingcited by 5
  22. 2023Towards Generalizable Data Protection With Transferable Unlearnable Examplescited by 4
  23. 2024Data-Dependent Stability Analysis of Adversarial Trainingcited by 4
  24. 2025How Far Are We from True Unlearnability?cited by 4
  25. 2024PureGen: Universal Data Purification for Train-Time Poison Defense via Generative Model Dynamicscited by 3
  26. 2024BridgePure: Limited Protection Leakage Can Break Black-Box Data Protectioncited by 3
  27. 2024ExpShield: Safeguarding Web Text from Unauthorized Crawling and LLM Exploitationcited by 3
  28. 2025T2UE: Generating Unlearnable Examples from Text Descriptionscited by 3
  29. 2026Full spectrum Unlearnable Examples via Spectral Equalizationcited by 3
  30. 2024Effective and Robust Adversarial Training against Data and Label Corruptionscited by 2
  31. 2025Provable Watermarking for Data Poisoning Attackscited by 2
  32. 2026Why Do Unlearnable Examples Work: A Novel Perspective of Mutual Informationcited by 2
  33. 2026Adversarial Attacks for Good: A Survey of Proactive Protection across the Visual Content Lifecyclecited by 2
  34. 2023Unlearnable Examples Give a False Sense of Data Privacy: Understanding and Relearningcited by 1
  35. 2024Be Persistent: Towards a Unified Solution for Mitigating Shortcuts in Deep Learningcited by 1
  36. 2024Beyond Full Poisoning: Effective Availability Attacks with Partial Perturbationcited by 1
  37. 2024Exploiting Supervised Poison Vulnerability to Strengthen Self-Supervised Defensecited by 1
  38. 2024Learning from Convolution-based Unlearnable Datasetscited by 1

50 most cited of 62 All 62 (JSON)