mlsanityData to 14 Oct 2026

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective

arXiv 2311.16646 ↗First posted 28 Nov 2023cs.LG cs.CRcited by 10

On the map: LLM safety and security → Backdoor and data poisoning attacks

Abstract (arXiv)

Dataset distillation offers a potential means to enhance data efficiency in deep learning. Recent studies have shown its ability to counteract backdoor risks present in original training samples. In this study, we delve into the theoretical aspects of backdoor attacks and dataset distillation based on kernel methods. We introduce two new theory-driven trigger pattern generation methods specialized for dataset distillation. Following a comprehensive set of analyses and experiments, we show that our optimization-based trigger design framework informs effective backdoor attacks on dataset distillation. Notably, datasets poisoned by our designed trigger prove resilient against conventional backdoor attack detection and mitigation methods. Our empirical results validate that the triggers developed using our approaches are proficient at executing resilient backdoor attacks.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security. Builds on 32 of its 44 references that are in the map. First posted on arXiv 28 Nov 2023.

Builds on

32 of 44 references.

Closest ancestors · 1

Same direction, newest first

  1. 2021Sleeper Agent: Scalable Hidden Trigger Backdoors for Neural Networks Trained from Scratch16 Jun

Other references · 31

  1. 2018Neural Tangent Kernel: Convergence and Generalization in Neural Networkscited by 2,621
  2. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
  3. 2019Wide Neural Networks of Any Depth Evolve as Linear Models Under Gradient Descentcited by 738
  4. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 580
  5. 2020Dataset Condensation with Gradient Matchingcited by 516
  6. 2018Spectral Signatures in Backdoor Attackscited by 484
  7. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 459
  8. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
  9. 2021Dataset Condensation with Distribution Matchingcited by 365
  10. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 353
  11. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 349
  12. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 324
19 more
  1. 2020Input-Aware Dynamic Backdoor Attackcited by 313
  2. 2019Label-Consistent Backdoor Attackscited by 296
  3. 2021Anti-Backdoor Learning: Training Clean Models on Poisoned Datacited by 263
  4. 2019A new Backdoor Attack in CNNs by training set corruption without label poisoningcited by 242
  5. 2020Dataset Meta-Learning from Kernel Ridge-Regressioncited by 230
  6. 2022Dataset Distillation using Neural Feature Regressioncited by 177
  7. 2023Dataset Distillation: A Comprehensive Reviewcited by 127
  8. 2019Demon in the Variant: Statistical Analysis of DNNs for Robust Backdoor Contamination Detectioncited by 124
  9. 2022Privacy for Free: How does Dataset Condensation Help Privacy?cited by 120
  10. 2022Dataset Distillation via Factorizationcited by 115
  11. 2022Efficient Dataset Distillation Using Random Feature Approximationcited by 111
  12. 2022Dataset Condensation with Contrastive Signalscited by 105
  13. 2023DREAM: Efficient Dataset Distillation by Representative Matchingcited by 94
  14. 2020Bayesian Deep Ensembles via the Neural Tangent Kernelcited by 90
  15. 2023Dataset Distillation with Convexified Implicit Gradientscited by 46
  16. 2023DiM: Distilling Dataset into Generative Modelcited by 46
  17. 2021Reproducing Kernel Hilbert Space, Mercer's Theorem, Eigenfunctions, Nystr"om Method, and Use of Kernels in Machine Learning: Tutorial and Surveycited by 37
  18. 2022Dataset Condensation with Latent Space Knowledge Factorization and Sharingcited by 34
  19. 2023Backdoor Attacks Against Dataset Distillationcited by 21

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 4; 2026: 3.

10 papers.

  1. 2025Dataset Distillation with Neural Characteristic Function: A Minmax Perspectivecited by 45
  2. 2025The Evolution of Dataset Distillation: Toward Scalable and Generalizable Solutionscited by 26
  3. 2024Not All Samples Should Be Utilized Equally: Towards Understanding and Improving Dataset Distillationcited by 14
  4. 2024Backdoor Graph Condensationcited by 5
  5. 2024FairDD: Fair Dataset Distillationcited by 3
  6. 2025Model Reprogramming Demystified: A Neural Tangent Kernel Perspectivecited by 2
  7. 2025DiRe: Diversity-promoting Regularization for Dataset Condensationcited by 1
  8. 2026InkDrop: Invisible Backdoor Attacks Against Dataset Condensationcited by 0
  9. 2026SNEAKDOOR: Stealthy Backdoor Attacks against Distribution Matching-based Dataset Condensationcited by 0
  10. 2026Dataset Distillation Based on Saliency-Driven Prototype Alignmentcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Dark Distillation: Backdooring Distilled Datasets without Accessing Raw Data
  2. 2024A Practical Trigger-Free Backdoor Attack on Neural Networks ↗
  3. 2021Rethinking the Backdoor Attacks' Triggers: A Frequency Perspective
  4. 2024Unlearn to Relearn Backdoors: Deferred Backdoor Functionality Attacks on Deep Learning Models
  5. 2025DeBackdoor: A Deductive Framework for Detecting Backdoor Attacks on Deep Models with Limited Data
  6. 2024Shortcuts Everywhere and Nowhere: Exploring Multi-Trigger Backdoor Attacks
  7. 2025BackWeak: Backdooring Knowledge Distillation Simply with Weak Triggers and Fine-tuning
  8. 2025How to Backdoor Image Knowledge Distillation