mlsanityData to 14 Oct 2026

Demystifying Poisoning Backdoor Attacks from a Statistical Perspective

arXiv 2310.10780 ↗First posted 16 Oct 2023cs.CR cs.AI cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

The growing dependence on machine learning in real-world applications emphasizes the importance of understanding and ensuring its safety. Backdoor attacks pose a significant security risk due to their stealthy nature and potentially serious consequences. Such attacks involve embedding triggers within a learning model with the intention of causing malicious behavior when an active trigger is present while maintaining regular functionality without it. This paper evaluates the effectiveness of any backdoor attack incorporating a constant trigger, by establishing tight lower and upper boundaries for the performance of the compromised model on both clean and backdoor test data. The developed theory answers a series of fundamental but previously underexplored problems, including (1) what are the determining factors for a backdoor attack's success, (2) what is the direction of the most effective backdoor attack, and (3) when will a human-imperceptible trigger succeed. Our derived understanding applies to both discriminative and generative models. We also demonstrate the theory by conducting experiments using benchmark datasets and state-of-the-art backdoor attack scenarios.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Builds on 37 of its 60 references that are in the map. First posted on arXiv 16 Oct 2023.

Builds on

37 of 60 references.

References · 37

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  5. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
  6. 2018How To Backdoor Federated Learningcited by 765
  7. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 580
  8. 2018Spectral Signatures in Backdoor Attackscited by 484
  9. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 459
  10. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
  11. 2020Backdoor Learning: A Surveycited by 426
  12. 2019A Survey of Deep Learning Techniques for Autonomous Drivingcited by 403
25 more
  1. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 353
  2. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 349
  3. 2020Input-Aware Dynamic Backdoor Attackcited by 313
  4. 2019A new Backdoor Attack in CNNs by training set corruption without label poisoningcited by 242
  5. 2020Blind Backdoors in Deep Learning Modelscited by 182
  6. 2020Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defensescited by 180
  7. 2020Clean-Label Backdoor Attacks on Video Recognition Modelscited by 169
  8. 2018SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systemscited by 152
  9. 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 130
  10. 2019Demon in the Variant: Statistical Analysis of DNNs for Robust Backdoor Contamination Detectioncited by 124
  11. 2022How to Backdoor Diffusion Models?cited by 104
  12. 2021Sleeper Agent: Scalable Hidden Trigger Backdoors for Neural Networks Trained from Scratchcited by 102
  13. 2022A Unified Evaluation of Textual Backdoor Learning: Frameworks and Benchmarkscited by 74
  14. 2018Model Selection Techniques -- An Overviewcited by 36
  15. 2022Identifying a Training-Set Attack's Target Using Renormalized Influence Estimationcited by 32
  16. 2023Pruning Deep Neural Networks from a Sparsity Perspectivecited by 25
  17. 2021Excess Capacity and Backdoor Poisoningcited by 24
  18. 2023Backdoor Learning on Sequence to Sequence Modelscited by 21
  19. 2022Federated Learning Challenges and Opportunities: An Outlookcited by 19
  20. 2022Data-Efficient Backdoor Attackscited by 13
  21. 2022A temporal chrominance trigger for clean-label backdoor attack against anti-spoof rebroadcast detectioncited by 10
  22. 2021GAL: Gradient Assisted Learning for Decentralized Multi-Organization Collaborationscited by 8
  23. 2019Restricted Recurrent Neural Networkscited by 5
  24. 2021The Rate of Convergence of Variation-Constrained Deep Neural Networkscited by 4
  25. 2023Provable Identifiability of Two-Layer ReLU Neural Networks via LASSO Regularizationcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 4; 2025: 5; 2026: 1.

10 papers.

  1. 2024DisDet: Exploring Detectability of Backdoor Attack on Diffusion Modelscited by 17
  2. 2024On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domainscited by 14
  3. 2025Probe Pruning: Accelerating LLMs through Dynamic Pruning via Model-Probingcited by 14
  4. 2024RAW: A Robust and Agile Plug-and-Play Watermark Framework for AI-Generated Images with Provable Guaranteescited by 8
  5. 2025An Outlook on the Opportunities and Challenges of Multi-Agent AI Systemscited by 8
  6. 2025Statistical Collusion by Collectives on Learning Platformscited by 6
  7. 2024iBA: Backdoor Attack on 3D Point Cloud via Reconstructing Itselfcited by 4
  8. 2025Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Datacited by 4
  9. 2025Provable one-poison backdoor attacks on linear models and ReLU neural networkscited by 0
  10. 2026Detecting and Eliminating Neural Network Backdoors Through Active Paths with Application to Intrusion Detectioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2021Backdoor Learning Curves: Explaining Backdoor Poisoning Beyond Influence Functions
  2. 2023Rethinking Backdoor Attacks
  3. 2023Stealthy Backdoor Attack via Confidence-driven Sampling
  4. 2023Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
  5. 2024Rethinking Backdoor Detection Evaluation for Language Models
  6. 2022Towards A Proactive ML Approach for Detecting Backdoor Poison Samples
  7. 2024Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor
  8. 2023Boosting Backdoor Attack with A Learnable Poisoning Sample Selection Strategy