mlsanityData to 13 Oct 2026

A Closer Look at Machine Unlearning for Large Language Models

arXiv 2410.08109 ↗First posted 10 Oct 2024cs.CL cs.AI cs.LGcited by 55

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

Large language models (LLMs) may memorize sensitive or copyrighted content, raising privacy and legal concerns. Due to the high cost of retraining from scratch, researchers attempt to employ machine unlearning to remove specific content from LLMs while preserving the overall performance. In this paper, we discuss several issues in machine unlearning for LLMs and provide our insights on possible approaches. To address the issue of inadequate evaluation of model outputs after unlearning, we introduce three additional metrics to evaluate token diversity, sentence semantics, and factual correctness. We then categorize unlearning methods into untargeted and targeted, and discuss their issues respectively. Specifically, the behavior that untargeted unlearning attempts to approximate is unpredictable and may involve hallucinations, and existing regularization is insufficient for targeted unlearning. To alleviate these issues, we propose using the objective of maximizing entropy (ME) for untargeted unlearning and incorporate answer preservation (AP) loss as regularization for targeted unlearning. Experimental results across three scenarios, i.e., fictitious unlearning, continual unlearning, and real-world unlearning, demonstrate the effectiveness of our approaches. The code is available at https://github.com/sail-sg/closer-look-LLM-unlearning.

Key facts. Cited by 55 papers in the map (data through 13 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 51 of its 65 references that are in the map. First posted on arXiv 10 Oct 2024.

Builds on

51 of 65 references.

Closest ancestors · 27

Same direction, newest first

  1. 2024Learn while Unlearn: An Iterative Unlearning Framework for Generative Language Models25 Jul
  2. 2024Learning to Refuse: Towards Mitigating Privacy Risks in LLMs14 Jul
  3. 2024MUSE: Machine Unlearning Six-Way Evaluation for Language Models8 Jul
  4. 2024To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models2 Jul
  5. 2024Soft Prompting for Unlearning in Large Language Models17 Jun
  6. 2024RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models16 Jun
  7. 2024Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference12 Jun
  8. 2024Large Language Model Unlearning via Embedding-Corrupted Prompts12 Jun
  9. 2024RKLD: Reverse KL-Divergence-based Knowledge Distillation for Unlearning Personal Information in Large Language Models4 Jun
  10. 2024SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning28 Apr
  11. 2024Offset Unlearning for Large Language Models17 Apr
  12. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning8 Apr
15 more
  1. 2024Second-Order Information Matters: Revisiting Machine Unlearning for Large Language Models13 Mar
  2. 2024Guardrail Baselines for Unlearning in LLMs5 Mar
  3. 2024Machine Unlearning of Pre-trained Large Language Models23 Feb
  4. 2024Towards Safer Large Language Models through Machine Unlearning15 Feb
  5. 2024Rethinking Machine Unlearning for Large Language Models13 Feb
  6. 2024Selective Forgetting: Advancing Machine Unlearning Techniques and Evaluation in Language Models8 Feb
  7. 2024Fast Exact Unlearning for In-Context Learning Data for LLMs1 Feb
  8. 2024TOFU: A Task of Fictitious Unlearning for LLMs11 Jan
  9. 2023Knowledge Unlearning for LLMs: Tasks, Methods, and Challenges27 Nov
  10. 2023Large Language Model Unlearning14 Oct
  11. 2023In-Context Unlearning: Language Models as Few Shot Unlearners11 Oct
  12. 2023Who's Harry Potter? Approximate Unlearning in LLMs3 Oct
  13. 2023Right to be Forgotten in the Era of Large Language Models: Implications, Challenges, and Solutions8 Jul
  14. 2022Knowledge Unlearning for Mitigating Privacy Risks in Language Models4 Oct
  15. 2022A Survey of Machine Unlearning6 Sep

Other references · 24

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,916
  2. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,107
  3. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,986
  4. 2020Measuring Massive Multitask Language Understandingcited by 7,876
  5. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,798
  6. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,949
  7. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,925
  8. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,594
  9. 2019Machine Unlearningcited by 1,056
  10. 2017SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluationcited by 998
  11. 2022Quantifying Memorization Across Neural Language Modelscited by 763
  12. 2024The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearningcited by 498
12 more
  1. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 331
  2. 2022Deduplicating Training Data Mitigates Privacy Risks in Language Modelscited by 314
  3. 2022Are Large Pre-Trained Language Models Leaking Your Personal Information?cited by 200
  4. 2023Beyond Memorization: Violating Privacy Via Inference with Large Language Modelscited by 173
  5. 2024Tamper-Resistant Safeguards for Open-Weight LLMscited by 136
  6. 2018Generating Informative and Diverse Conversational Responses via Adversarial Information Maximizationcited by 115
  7. 2022Preventing Verbatim Memorization in Language Models Gives a False Sense of Privacycited by 99
  8. 2024Eraser: Jailbreaking Defense in Large Language Models via Unlearning Harmful Knowledgecited by 49
  9. 2024Instruction Tuning With Loss Over Instructionscited by 40
  10. 2023Knowledge Sanitization of Large Language Modelscited by 36
  11. 2024From Theft to Bomb-Making: The Ripple Effect of Unlearning in Defending Against Jailbreak Attackscited by 29
  12. 2024Accurate and Nuanced Open-QA Evaluation Through Textual Entailmentcited by 12

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 27; 2026: 28.

55 papers.

Sort:

By area:

  1. 2025Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMscited by 41
  2. 2025A Comprehensive Survey of Machine Unlearning Techniques for Large Language Modelscited by 32
  3. 2025Security Concerns for Large Language Models: A Surveycited by 29
  4. 2025CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuningcited by 11
  5. 2025Constrained Entropic Unlearning: A Primal-Dual Framework for Large Language Modelscited by 10
  6. 2025SoK: Machine Unlearning for Large Language Modelscited by 10
  7. 2025Leak@k: Unlearning Does Not Make LLMs Forget Under Probabilistic Decodingcited by 9
  8. 2025Improving LLM Unlearning Robustness via Random Perturbationscited by 7
  9. 2025AI Supply Chains: An Emerging Ecosystem of AI Actors, Products, and Servicescited by 7
  10. 2025DualOptim: Enhancing Efficacy and Stability in Machine Unlearning with Dual Optimizerscited by 6
  11. 2025Step-by-Step Reasoning Attack: Revealing 'Erased' Knowledge in Large Language Modelscited by 6
  12. 2026Attention Smoothing Is All You Need For Unlearningcited by 6
38 more
  1. 2025UniErase: Towards Balanced and Precise Unlearning in Language Modelscited by 5
  2. 2025Proximalized Preference Optimization for Diverse Feedback Types: A Decomposed Perspective on DPOcited by 5
  3. 2025LLM Unlearning Without an Expert Curated Datasetcited by 5
  4. 2025Wisdom is Knowing What not to Say: Hallucination-Free LLMs Unlearning via Attention Shiftingcited by 5
  5. 2025Curvature-Aware Safety Restoration In LLMs Fine-Tuningcited by 5
  6. 2025Forgetting: A New Mechanism Towards Better Large Language Model Fine-tuningcited by 4
  7. 2025Understanding the Dilemma of Unlearning for Large Language Modelscited by 4
  8. 2026Reinforcement Unlearning via Group Relative Policy Optimizationcited by 4
  9. 2026REBEL: Hidden Knowledge Recovery via Evolutionary-Based Evaluation Loopcited by 4
  10. 2025GRAIL: Gradient-Based Adaptive Unlearning for Privacy and Copyright in LLMscited by 3
  11. 2026Position: The Term "Machine Unlearning" Is Overused in LLMscited by 3
  12. 2025Direct Token Optimization: A Self-contained Approach to Large Language Model Unlearningcited by 2
  13. 2026STaR: Sensitive Trajectory Regulation for Unlearning in Large Reasoning Modelscited by 2
  14. 2026Multi-Head Attention Is a Multi-Player Gamecited by 2
  15. 2025MIAT: Maneuver-Intention-Aware Transformer for Spatio-Temporal Trajectory Predictioncited by 1
  16. 2025Unlearning at Scale: State-Exact Trace-Preserving Deletion in Billion-Parameter Language Modelscited by 1
  17. 2025Standard vs. Modular Sampling: Best Practices for Reliable LLM Unlearningcited by 1
  18. 2025LLM Unlearning Under the Microscope: A Full-Stack View on Methods and Metricscited by 1
  19. 2026Beyond Forgetting: Representation Misdirection Elicits Controllable Side Behaviors and Capabilitiescited by 1
  20. 2026Rethinking Benign Relearning: Syntax as the Hidden Driver of Unlearning Failurescited by 1
  21. 2026KUDA: Knowledge Unlearning by Deviating Representation for Large Language Modelscited by 1
  22. 2026Which Concepts to Forget and How to Refuse? Decomposing Concepts for Continual Unlearning in Large Vision-Language Modelscited by 1
  23. 2026Secure Forgetting: A Framework for Privacy-Driven Unlearning in Large Language Model (LLM)-Based Agentscited by 1
  24. 2026Subspace Control: Turning Constrained Model Steering into Controllable Spectral Optimizationcited by 1
  25. 2026Position: AI Safety Requires Effective Controllabilitycited by 1
  26. 2026Modular Pretraining Enables Access Controlcited by 1
  27. 2026Beyond Binary Rewards: A Comparative Study of Reward Design for Reinforcement Unlearningcited by 1
  28. 2025Lifting Data-Tracing Machine Unlearning to Knowledge-Tracing for Foundation Modelscited by 0
  29. 2025Controllable Machine Unlearning via Gradient Pivotingcited by 0
  30. 2025RapidUn: Influence-Driven Parameter Reweighting for Efficient Large Language Model Unlearningcited by 0
  31. 2026Shadow Unlearning: A Neuro-Semantic Approach to Fidelity-Preserving Faceless Forgetting in LLMscited by 0
  32. 2026FIT to Forget: Robust Continual Unlearning for Large Language Modelscited by 0
  33. 2026EvoMU: Evolutionary Machine Unlearningcited by 0
  34. 2026Exclusive Unlearningcited by 0
  35. 2026Machine Unlearning for Masked Diffusion Language Modelscited by 0
  36. 2026DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language Modelscited by 0
  37. 2026Stochastic Meta-Unlearning: Bridging Language Backbone and Multimodal Unlearningcited by 0
  38. 2026Spectral Saliency for Machine Unlearningcited by 0

50 most cited of 55 All 55 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Unlearning in LLMs: Methods, Evaluation, and Open Challenges
  2. 2024Catastrophic Failure of LLM Unlearning via Quantization
  3. 2024Machine Unlearning in Large Language Models
  4. 2025A Survey on Unlearning in Large Language Models
  5. 2025Unlearning That Lasts: Utility-Preserving, Robust, and Almost Irreversible Forgetting in LLMs
  6. 2025Not Every Token Needs Forgetting: Selective Unlearning to Limit Change in Utility in Large Language Model Unlearning
  7. 2025Scalable and Robust LLM Unlearning by Correcting Responses with Retrieved Exclusions
  8. 2024Towards Effective Evaluations and Comparisons for LLM Unlearning Methods