mlsanityData to 7 Oct 2026

Backdoor Defense with Machine Unlearning

arXiv 2201.09538 ↗First posted 24 Jan 2022cs.CR cs.LGcited by 61

Not on the map. Why?

Cited by 45 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Backdoor injection attack is an emerging threat to the security of neural networks, however, there still exist limited effective defense methods against the attack. In this paper, we propose BAERASE, a novel method that can erase the backdoor injected into the victim model through machine unlearning. Specifically, BAERASE mainly implements backdoor defense in two key steps. First, trigger pattern recovery is conducted to extract the trigger patterns infected by the victim model. Here, the trigger pattern recovery problem is equivalent to the one of extracting an unknown noise distribution from the victim model, which can be easily resolved by the entropy maximization based generative model. Subsequently, BAERASE leverages these recovered trigger patterns to reverse the backdoor injection procedure and induce the victim model to erase the polluted memories through a newly designed gradient ascent based machine unlearning method. Compared with the previous machine unlearning solutions, the proposed approach gets rid of the reliance on the full access to training data for retraining and shows higher effectiveness on backdoor erasing than existing fine-tuning or pruning methods. Moreover, experiments show that BAERASE can averagely lower the attack success rates of three kinds of state-of-the-art backdoor attacks by 99% on four benchmark datasets.

Key facts. Cited by 61 papers in the map (data through 7 Oct 2026). Builds on 17 of its 28 references that are in the map. First posted on arXiv 24 Jan 2022.

Builds on

17 of 28 references.

References · 17

  1. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  2. 2019Machine Unlearningcited by 1,020
  3. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
  4. 2018Spectral Signatures in Backdoor Attackscited by 457
  5. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 437
  6. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 313
  7. 2020Descent-to-Delete: Gradient-Based Methods for Machine Unlearningcited by 303
  8. 2018SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systemscited by 148
  9. 2019Universal Litmus Patterns: Revealing Backdoor Attacks in CNNscited by 143
  10. 2018Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbationcited by 131
  11. 2021Graph Unlearningcited by 129
  12. 2020Towards Robust LiDAR-based Perception in Autonomous Driving: General Black-box Adversarial Sensor Attack and Countermeasurescited by 95
5 more
  1. 2019Robust Anomaly Detection and Backdoor Attack Detection Via Differential Privacycited by 93
  2. 2020Practical Detection of Trojan Neural Networks: Data-Limited and Data-Free Casescited by 93
  3. 2019Defending Neural Backdoors via Generative Distribution Modelingcited by 82
  4. 2020One-pixel Signature: Characterizing CNN Models for Backdoor Detectioncited by 34
  5. 2019Revocable Federated Learning: A Benchmark of Federated Forestcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 7; 2023: 11; 2024: 22; 2025: 14; 2026: 7.

61 papers.

Sort:

By area:

  1. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 376
  2. 2022A Survey of Machine Unlearningcited by 298
  3. 2023Large Language Model Unlearningcited by 288
  4. 2024Rethinking Machine Unlearning for Large Language Modelscited by 247
  5. 2023Model Sparsity Can Simplify Machine Unlearningcited by 206
  6. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Nowcited by 204
  7. 2022Federated Unlearning: How to Efficiently Erase a Client in FL?cited by 90
  8. 2023Machine Unlearning: Solutions and Challengescited by 85
  9. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 66
  10. 2024UnlearnCanvas: Stylized Image Dataset for Enhanced Machine Unlearning Evaluation in Diffusion Modelscited by 47
  11. 2024UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AIcited by 46
  12. 2024Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearningcited by 44
38 more
  1. 2023GNNDelete: A General Strategy for Unlearning in Graph Neural Networkscited by 43
  2. 2022Evaluating Machine Unlearning via Epistemic Uncertaintycited by 39
  3. 2022FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learningcited by 37
  4. 2024A Survey of Recent Backdoor Attacks and Defenses in Large Language Modelscited by 36
  5. 2024Threats, Attacks, and Defenses in Machine Unlearning: A Surveycited by 34
  6. 2024Machine Unlearning: A Comprehensive Surveycited by 33
  7. 2024A Review on Machine Unlearningcited by 30
  8. 2025Decoupled Distillation to Erase: A General Unlearning Method for Any Class-centric Taskscited by 28
  9. 2024Corrective Machine Unlearningcited by 26
  10. 2024TERD: A Unified Framework for Safeguarding Diffusion Models Against Backdoorscited by 25
  11. 2024WAGLE: Strategic Weight Attribution for Effective and Modular Unlearning in Large Language Modelscited by 25
  12. 2022BAFFLE: Hiding Backdoors in Offline Reinforcement Learning Datasetscited by 24
  13. 2025SAEs Can Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMscited by 21
  14. 2023Machine Unlearning: A Surveycited by 18
  15. 2022Provable Defense against Backdoor Policies in Reinforcement Learningcited by 16
  16. 2023Get Rid Of Your Trail: Remotely Erasing Backdoors in Federated Learningcited by 12
  17. 2023A Survey of Graph Unlearningcited by 12
  18. 2024UPAM: Unified Prompt Attack in Text-to-Image Generation Models Against Both Textual Filters and Visual Checkerscited by 11
  19. 2024Towards Unified Robustness Against Both Backdoor and Adversarial Attackscited by 9
  20. 2024Label Smoothing Improves Machine Unlearningcited by 8
  21. 2024Semantic Shield: Defending Vision-Language Models Against Backdooring and Poisoning via Fine-grained Knowledge Alignmentcited by 7
  22. 2024A Survey on Recommendation Unlearning: Fundamentals, Taxonomy, Evaluation, and Open Questionscited by 7
  23. 2024CURE4Rec: A Benchmark for Recommendation Unlearning with Deeper Influencecited by 6
  24. 2023Machine Unlearning: its nature, scope, and importance for a "delete culture"cited by 5
  25. 2025Unified Prompt Attack Against Text-to-Image Generation Modelscited by 5
  26. 2025Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!cited by 5
  27. 2024Mudjacking: Patching Backdoor Vulnerabilities in Foundation Modelscited by 4
  28. 2025Forgetting to Forget: Attention Sink as A Gateway for Backdooring LLM Unlearningcited by 4
  29. 2024BlockFUL: Enabling Unlearning in Blockchained Federated Learningcited by 3
  30. 2024Defending Deep Regression Models against Backdoor Attackscited by 3
  31. 2025SineProject: Machine Unlearning for Stable Vision Language Alignmentcited by 3
  32. 2026Is your algorithm unlearning or untraining?cited by 3
  33. 2022Challenges and Pitfalls of Bayesian Unlearningcited by 2
  34. 2025Bad-PFL: Exploring Backdoor Attacks against Personalized Federated Learningcited by 2
  35. 2026Patcher: Post-Hoc Patching of Backdoored Large Language Modelscited by 2
  36. 2024Dataset Condensation Driven Machine Unlearningcited by 1
  37. 2024Unlearning Backdoor Attacks through Gradient-Based Model Pruningcited by 1
  38. 2025DISTIL: Data-Free Inversion of Suspicious Trojan Inputs via Latent Diffusioncited by 1

50 most cited of 61 All 61 (JSON)