mlsanityData to 7 Oct 2026

A Survey of Machine Unlearning

arXiv 2209.02299 ↗First posted 6 Sep 2022cs.LG cs.AIcited by 298

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

Today, computer systems hold large amounts of personal data. Yet while such an abundance of data allows breakthroughs in artificial intelligence, and especially machine learning (ML), its existence can be a threat to user privacy, and it can weaken the bonds of trust between humans and AI. Recent regulations now require that, on request, private information about a user must be removed from both computer systems and from ML models, i.e. ``the right to be forgotten''). While removing data from back-end databases should be straightforward, it is not sufficient in the AI context as ML models often `remember' the old data. Contemporary adversarial attacks on trained models have proven that we can learn whether an instance or an attribute belonged to the training data. This phenomenon calls for a new paradigm, namely machine unlearning, to make ML models forget about particular data. It turns out that recent works on machine unlearning have not been able to completely solve the problem due to the lack of common frameworks and resources. Therefore, this paper aspires to present a comprehensive examination of machine unlearning's concepts, scenarios, methods, and applications. Specifically, as a category collection of cutting-edge studies, the intention behind this article is to serve as a comprehensive resource for researchers and practitioners seeking an introduction to machine unlearning and its formulations, design criteria, removal requests, algorithms, and applications. In addition, we aim to highlight the key findings, current trends, and new research areas that have not yet featured the use of machine unlearning but could benefit greatly from it. We hope this survey serves as a valuable resource for ML researchers and those seeking to innovate privacy technologies. Our resources are publicly available at https://github.com/tamlhp/awesome-machine-unlearning.

Key facts. Cited by 298 papers in the map (data through 7 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 107 of its 263 references that are in the map. First posted on arXiv 6 Sep 2022.

Builds on

107 of 263 references.

Closest ancestors · 25

Same direction, newest first

  1. 2022An Introduction to Machine Unlearning2 Sep
  2. 2022Evaluating Machine Unlearning via Epistemic Uncertainty23 Aug
  3. 2022Few-Shot Unlearning by Model Inversion31 May
  4. 2022Quark: Controllable Text Generation with Reinforced Unlearning26 May
  5. 2022Can Bad Teaching Induce Forgetting? Unlearning in Deep Networks using an Incompetent Teacher17 May
  6. 2022Deep Unlearning via Randomized Conditionally Independent Hessians15 Apr
  7. 2022Knowledge Removal in Sampling-based Bayesian Inference24 Mar
  8. 2022Continual Learning and Private Unlearning24 Mar
  9. 2022Markov Chain Monte Carlo-Based Machine Unlearning: Unlearning What Needs to be Forgotten28 Feb
  10. 2022Efficient Attribute Unlearning: Towards Selective Removal of Input Attributes from Feature Representations27 Feb
  11. 2022Deletion Inference, Reconstruction, and Compliance in Machine (Un)Learning7 Feb
  12. 2022Zero-Shot Machine Unlearning14 Jan
13 more
  1. 2021An Investigation on Learning, Polluting, and Unlearning the Spam Emails for Lifelong Learning26 Nov
  2. 2021Machine unlearning via GAN22 Nov
  3. 2021Fast Yet Effective Machine Unlearning17 Nov
  4. 2021On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning22 Oct
  5. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearning27 Sep
  6. 2021Hard to Forget: Poisoning Attacks on Certified Machine Unlearning17 Sep
  7. 2021Machine Unlearning of Features and Labels26 Aug
  8. 2021Certifiable Machine Unlearning for Linear Models29 Jun
  9. 2021Adaptive Machine Unlearning8 Jun
  10. 2021A unified PAC-Bayesian framework for machine unlearning via information risk minimization1 Jun
  11. 2021DeepObliviate: A Powerful Charm for Erasing Data Residual Memory in Deep Neural Networks13 May
  12. 2021Remember What You Want to Forget: Algorithms for Machine Unlearning4 Mar
  13. 2021Machine Unlearning via Algorithmic Stability25 Feb

Other references · 82

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  3. 2016Deep Learning with Differential Privacycited by 2,682
  4. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  5. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  6. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
  7. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  8. 2019Simplifying Graph Convolutional Networkscited by 1,470
  9. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  10. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,313
  11. 2019Machine Unlearningcited by 1,020
  12. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 903
38 more
  1. 2015Deep Learning and the Information Bottleneck Principlecited by 793
  2. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 599
  3. 2021Knowledge Neurons in Pretrained Transformerscited by 590
  4. 2019Certified Data Removal from Machine Learning Modelscited by 512
  5. 2017Opening the Black Box of Deep Neural Networks via Informationcited by 494
  6. 2018ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Modelscited by 475
  7. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 447
  8. 2014New insights and perspectives on the natural gradient methodcited by 435
  9. 2020Amnesiac Machine Learningcited by 305
  10. 2020Descent-to-Delete: Gradient-Based Methods for Machine Unlearningcited by 303
  11. 2018Deep Face Recognition: A Surveycited by 275
  12. 2019Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigationcited by 252
  13. 2020Approximate Data Deletion from Machine Learning Modelscited by 250
  14. 2020Influence Functions in Deep Learning Are Fragilecited by 247
  15. 2021Unlearnable Examples: Making Personal Data Unexploitablecited by 188
  16. 2020When Machine Unlearning Jeopardizes Privacycited by 181
  17. 2020Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observationscited by 171
  18. 2020Mixed-Privacy Forgetting in Deep Networkscited by 158
  19. 2020DeltaGrad: Rapid retraining of machine learning modelscited by 143
  20. 2019Zero-shot Knowledge Transfer via Adversarial Belief Matchingcited by 129
  21. 2021Graph Unlearningcited by 129
  22. 2020Machine Unlearning for Random Forestscited by 125
  23. 2020Fair Attribute Classification through Latent Space De-biasingcited by 125
  24. 2022Measuring Forgetting of Memorized Training Examplescited by 118
  25. 2019Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learningcited by 117
  26. 2021Distilling Causal Effect of Data in Class-Incremental Learningcited by 117
  27. 2020Variational Bayesian Unlearningcited by 114
  28. 2022The Right to be Forgotten in Federated Learning: An Efficient Realization with Rapid Retrainingcited by 109
  29. 2021Federated Unlearning via Class-Discriminative Pruningcited by 108
  30. 2018Uncertainty in Neural Networks: Approximately Bayesian Ensemblingcited by 94
  31. 2022Federated Unlearning: How to Efficiently Erase a Client in FL?cited by 90
  32. 2019Analyzing Information Leakage of Updates to Natural Language Modelscited by 89
  33. 2020Machine Unlearning: Linear Filtration for Logit-based Classifierscited by 89
  34. 2021Proof-of-Learning: Definitions and Practicecited by 81
  35. 2019Graph Representation Learning: A Surveycited by 75
  36. 2022Recommendation Unlearningcited by 71
  37. 2022Federated Unlearning with Knowledge Distillationcited by 70
  38. 2021FedAUX: Leveraging Unlabeled Auxiliary Data in Federated Learningcited by 67

50 most cited of 82 All 82 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 42; 2024: 95; 2025: 79; 2026: 80.

298 papers.

Sort:

By area:

  1. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearningcited by 484
  2. 2023Cognitive Architectures for Language Agentscited by 422
  3. 2024Mechanistic Interpretability for AI Safety -- A Reviewcited by 415
  4. 2023Who's Harry Potter? Approximate Unlearning in LLMscited by 366
  5. 2024HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Modelscited by 273
  6. 2024MUSE: Machine Unlearning Six-Way Evaluation for Language Modelscited by 251
  7. 2024Rethinking Machine Unlearning for Large Language Modelscited by 247
  8. 2023Model Sparsity Can Simplify Machine Unlearningcited by 206
  9. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Nowcited by 204
  10. 2023Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampeningcited by 177
  11. 2024Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Modelscited by 164
  12. 2024Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearningcited by 160
38 more
  1. 2024A Comprehensive Study of Knowledge Editing for Large Language Modelscited by 157
  2. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 155
  3. 2024Eight Methods to Evaluate Robust Unlearning in LLMscited by 153
  4. 2023Explainable Artificial Intelligence (XAI) 2.0: A Manifesto of Open Challenges and Interdisciplinary Research Directionscited by 123
  5. 2024Large Language Model Unlearning via Embedding-Corrupted Promptscited by 112
  6. 2024What makes unlearning hard and what to do about itcited by 89
  7. 2023Machine Unlearning: Solutions and Challengescited by 85
  8. 2025OpenUnlearning: Accelerating LLM Unlearning via Unified Benchmarking of Methods and Metricscited by 80
  9. 2024Safeguarding Large Language Models: A Surveycited by 73
  10. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 66
  11. 2024Do Unlearning Methods Remove Information from Language Model Weights?cited by 66
  12. 2024Navigating the Risks: A Survey of Security, Privacy, and Ethics Threats in LLM-Based Agentscited by 63
  13. 2024What Makes and Breaks Safety Fine-tuning? A Mechanistic Studycited by 54
  14. 2024A Closer Look at Machine Unlearning for Large Language Modelscited by 54
  15. 2024Scissorhands: Scrub Data Influence via Connection Sensitivity in Networkscited by 50
  16. 2024Are we making progress in unlearning? Findings from the first NeurIPS unlearning competitioncited by 48
  17. 2024Machine Unlearning for Image-to-Image Generative Modelscited by 47
  18. 2024UnlearnCanvas: Stylized Image Dataset for Enhanced Machine Unlearning Evaluation in Diffusion Modelscited by 47
  19. 2023Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomycited by 44
  20. 2024Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearningcited by 44
  21. 2023Reproducing Whisper-Style Training Using an Open-Source Toolkit and Publicly Available Datacited by 42
  22. 2024To Each (Textual Sequence) Its Own: Improving Memorized-Data Unlearning in Large Language Modelscited by 42
  23. 2024Dissecting Language Models: Machine Unlearning via Selective Pruningcited by 38
  24. 2025LLM Unlearning via Neural Activation Redirectioncited by 38
  25. 2022Privacy Adhering Machine Un-learning in NLPcited by 35
  26. 2024Copyright Protection in Generative AI: A Technical Perspectivecited by 35
  27. 2023A Survey on Knowledge Editing of Neural Networkscited by 34
  28. 2024Threats, Attacks, and Defenses in Machine Unlearning: A Surveycited by 34
  29. 2024Machine Unlearning: A Comprehensive Surveycited by 33
  30. 2022Deep Regression Unlearningcited by 32
  31. 2024Unlearning Backdoor Threats: Enhancing Backdoor Defense in Multimodal Contrastive Learning via Local Token Unlearningcited by 31
  32. 2024Continual Forgetting for Pre-trained Vision Modelscited by 29
  33. 2024Undesirable Memorization in Large Language Models: A Surveycited by 29
  34. 2024Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Datasetcited by 29
  35. 2024Towards Certified Unlearning for Deep Neural Networkscited by 28
  36. 2025Decoupled Distillation to Erase: A General Unlearning Method for Any Class-centric Taskscited by 28
  37. 2024Corrective Machine Unlearningcited by 26
  38. 2024Federated Unlearning: A Survey on Methods, Design Guidelines, and Evaluation Metricscited by 25

50 most cited of 298 All 298 (JSON)