pith. sign in

arxiv: 2405.07406 · v3 · submitted 2024-05-13 · 💻 cs.CR · cs.AI

Machine Unlearning: A Comprehensive Survey

Weiqi Wang , Zhiyi Tian , Chenhan Zhang , Shui Yu This is my paper

Pith reviewed 2026-05-24 01:11 UTC · model grok-4.3

classification 💻 cs.CR cs.AI
keywords machine unlearningright to be forgottenprivacy preservationfederated unlearninggraph unlearningunlearning verificationdata deletionmodel security
0
0 comments X

The pith

Machine unlearning methods divide into four scenarios: centralized, distributed, verification, and privacy-security.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The survey organizes techniques that remove specific data contributions from trained models to support privacy rights such as the right to be forgotten. It partitions the literature into centralized unlearning (split into exact and approximate approaches), distributed and irregular data unlearning (with federated and graph unlearning as examples), unlearning verification, and privacy and security issues. The classification highlights differences, connections, and open problems across these areas while detailing key techniques in each.

Core claim

Current machine unlearning methods are categorized into four scenarios: centralized unlearning (further split into exact and approximate), distributed and irregular data unlearning (represented by federated unlearning and graph unlearning), unlearning verification, and privacy and security issues in unlearning. The survey discusses their differences, connections, and open problems.

What carries the argument

The four-scenario taxonomy that partitions machine unlearning methods and organizes their techniques, differences, and challenges.

If this is right

  • Centralized unlearning splits into exact methods that fully retrain models without the target data and approximate methods that adjust parameters efficiently.
  • Distributed and irregular data cases require dedicated approaches for federated learning and graph-structured data.
  • Verification methods are required to confirm successful removal of data influence.
  • Privacy and security risks arise in the unlearning process itself and must be mitigated alongside the unlearning goals.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The taxonomy could support creation of standardized benchmarks that compare methods across scenarios.
  • Resolving verification challenges may enable enforceable standards for data deletion compliance in deployed models.
  • Security considerations indicate that unlearning operations could create new attack surfaces if not designed with those risks in mind.

Load-bearing premise

Existing machine unlearning literature partitions cleanly into the four proposed scenarios without substantial overlap, omission, or misclassification.

What would settle it

A substantial body of unlearning methods that cannot be assigned to any single scenario or that cross categories in ways that undermine the partitioning utility.

Figures

Figures reproduced from arXiv: 2405.07406 by Chenhan Zhang, Shui Yu, Weiqi Wang, Zhiyi Tian.

Figure 1
Figure 1. Figure 1: Our taxonomy for machine unlearning. The introduction order will also follow this figure. We classify the current [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the machine unlearning pipeline, from model training and unlearning request types to unlearning [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Privacy Leakage: a Privacy Reconstruction Process [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The model changes when adding a new point or removing a point. (a) A normally trained classifying model classifies [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: (a) Naive unlearning. There are only two steps: delete the specified samples from the whole dataset and retrain a [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The approximate unlearning by certified data removal. [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Client-level (Server-side) Federated Unlearning. Since the local data cannot be uploaded to the federated learning (FL) server side, most federated unlearning methods try to erase a certain client’s contribution from the trained model by storing and estimating the contribution of uploaded parameters. In this situation, they can implement federated unlearning without interacting with the client, shown as th… view at source ↗
Figure 7
Figure 7. Figure 7: Comparison between (a) server-side federated unlearning and (b) client-side federated unlearning [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: (a) Regular data unlearning, which only unlearns the data sample. (b) Graph unlearning. Since the graph data contains [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗
read the original abstract

As the right to be forgotten has been legislated worldwide, many studies attempt to design unlearning mechanisms to protect users' privacy when they want to leave machine learning service platforms. Specifically, machine unlearning is to make a trained model to remove the contribution of an erased subset of the training dataset. This survey aims to systematically classify a wide range of machine unlearning and discuss their differences, connections and open problems. We categorize current unlearning methods into four scenarios: centralized unlearning, distributed and irregular data unlearning, unlearning verification, and privacy and security issues in unlearning. Since centralized unlearning is the primary domain, we use two parts to introduce: firstly, we classify centralized unlearning into exact unlearning and approximate unlearning; secondly, we offer a detailed introduction to the techniques of these methods. Besides the centralized unlearning, we notice some studies about distributed and irregular data unlearning and introduce federated unlearning and graph unlearning as the two representative directions. After introducing unlearning methods, we review studies about unlearning verification. Moreover, we consider the privacy and security issues essential in machine unlearning and organize the latest related literature. Finally, we discuss the challenges of various unlearning scenarios and address the potential research directions.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper is a survey on machine unlearning motivated by privacy regulations such as the right to be forgotten. It claims to systematically classify existing work into four scenarios—centralized unlearning (subdivided into exact and approximate), distributed and irregular data unlearning (with federated and graph unlearning as representatives), unlearning verification, and privacy/security issues—while discussing differences, connections, open problems, and future directions. The abstract outlines the structure: detailed treatment of centralized methods, brief coverage of distributed/irregular cases, followed by separate reviews of verification and privacy/security topics.

Significance. If the taxonomy holds without substantial overlap or misclassification, the survey would provide a structured entry point into a rapidly growing literature, helping researchers identify connections between method families and ancillary topics such as verification. The explicit coverage of open problems and research directions is a positive contribution for guiding future work.

major comments (2)
  1. [Abstract] Abstract: The central claim is a classification of 'a wide range of machine unlearning' into four scenarios, yet the text explicitly separates the first two (centralized and distributed/irregular unlearning) as 'unlearning methods' and the last two as post-method reviews ('After introducing unlearning methods, we review studies about unlearning verification' and 'Moreover, we consider the privacy and security issues'). This produces a non-parallel taxonomy in which verification and privacy/security are not classes of unlearning techniques, undermining the utility and consistency of the proposed partitioning.
  2. [Abstract] Abstract and §1 (presumed introduction of the taxonomy): The weakest assumption—that the literature partitions cleanly into the four scenarios without substantial overlap or omission—is not supported by the abstract's own wording, which treats verification and privacy/security as ancillary rather than co-equal method scenarios. A concrete test would be whether papers on verification are ever presented as instances of 'unlearning methods' in the body; if not, the four-way scheme requires revision to maintain internal consistency.
minor comments (1)
  1. [Abstract] Abstract: The phrase 'distributed and irregular data unlearning' is used as a scenario label but then instantiated only by federated unlearning and graph unlearning; a brief justification for treating these two as representative of the broader category would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thoughtful analysis of our taxonomy and abstract. The comments highlight a genuine inconsistency in how the four scenarios are framed versus how the manuscript body structures the content. We address each point below and commit to revisions that improve clarity and consistency without altering the survey's core coverage.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim is a classification of 'a wide range of machine unlearning' into four scenarios, yet the text explicitly separates the first two (centralized and distributed/irregular unlearning) as 'unlearning methods' and the last two as post-method reviews ('After introducing unlearning methods, we review studies about unlearning verification' and 'Moreover, we consider the privacy and security issues'). This produces a non-parallel taxonomy in which verification and privacy/security are not classes of unlearning techniques, undermining the utility and consistency of the proposed partitioning.

    Authors: We agree that the abstract's wording creates a non-parallel structure. The body text does treat centralized and distributed/irregular unlearning as the core 'unlearning methods' sections, while verification and privacy/security are presented as subsequent reviews. This distinction was intended to reflect the literature's emphasis but was not clearly signaled in the taxonomy claim. We will revise the abstract (and corresponding introduction) to describe the survey as covering two primary unlearning method categories plus dedicated sections on verification and privacy/security issues, thereby aligning the high-level claim with the actual organization. revision: yes

  2. Referee: [Abstract] Abstract and §1 (presumed introduction of the taxonomy): The weakest assumption—that the literature partitions cleanly into the four scenarios without substantial overlap or omission—is not supported by the abstract's own wording, which treats verification and privacy/security as ancillary rather than co-equal method scenarios. A concrete test would be whether papers on verification are ever presented as instances of 'unlearning methods' in the body; if not, the four-way scheme requires revision to maintain internal consistency.

    Authors: The observation is accurate: verification papers are reviewed separately and are not framed in the body as instances of 'unlearning methods.' The four-scenario claim in the abstract therefore overstates the parallelism. We will revise the taxonomy statement to avoid implying a clean four-way partition of methods and instead present the categories as the main areas covered by the survey (methods in centralized and distributed settings, followed by verification and privacy/security analyses). This preserves the survey's utility while removing the internal inconsistency. revision: yes

Circularity Check

0 steps flagged

No circularity: descriptive survey with no derivations or predictions

full rationale

This is a literature survey paper whose central contribution is a proposed taxonomy of existing machine unlearning methods. No equations, fitted parameters, predictions, or first-principles derivations appear anywhere in the manuscript. The four-scenario classification is presented as an organizing framework chosen by the authors; it does not reduce to any self-referential definition, fitted input renamed as prediction, or load-bearing self-citation chain. The abstract and structure explicitly separate method reviews from subsequent verification and privacy sections, confirming the work is self-contained as a descriptive review without opportunity for the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a survey paper. It introduces no free parameters, axioms, or invented entities; all content is drawn from existing published work on machine unlearning.

pith-pipeline@v0.9.0 · 5748 in / 1038 out tokens · 31721 ms · 2026-05-24T01:11:30.754564+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Knowledge Beyond Language: Bridging the Gap in Multilingual Machine Unlearning Evaluation

    cs.CL 2026-05 unverdicted novelty 7.0

    New metrics KSS and KPS are introduced to evaluate multilingual machine unlearning quality and cross-language consistency in LLMs, addressing limitations of single-language evaluation protocols.

  2. Copyright Laundering Through the AI Ouroboros: Adapting the 'Fruit of the Poisonous Tree' Doctrine to Recursive AI Training

    cs.CY 2026-01 conditional novelty 7.0

    The paper introduces an AI-FOPT standard that presumes copyright infringement taint in models derived from an infringing foundational model unless developers prove independent lawful sourcing.

  3. Class Unlearning via Depth-Aware Removal of Forget-Specific Directions

    cs.CV 2026-04 unverdicted novelty 6.0

    DAMP performs one-shot class unlearning by extracting and projecting out forget-specific residual directions at each network depth using class prototypes and a separability-derived scaling rule.

  4. Class Unlearning via Depth-Aware Removal of Forget-Specific Directions

    cs.CV 2026-04 unverdicted novelty 6.0

    DAMP performs one-shot class unlearning by depth-aware projection removal of forget-specific directions, producing forgetting behavior closer to retraining from scratch than prior methods on image classification tasks.

  5. WIN-U: Woodbury-Informed Newton-Unlearning as a retain-free Machine Unlearning Framework

    cs.LG 2026-04 unverdicted novelty 6.0

    WIN-U delivers a retain-free unlearning update that approximates the gold-standard retrained model via a Woodbury-informed Newton step using only forget-set curvature information.

  6. Forget What's Sensitive, Remember What Matters: Token-Level Differential Privacy in Memory Sculpting for Continual Learning

    cs.AI 2025-09 unverdicted novelty 6.0

    PeCL applies token-level dynamic differential privacy and privacy-guided memory sculpting to achieve superior privacy-utility balance in continual learning.

  7. Towards Reliable Forgetting: A Survey on Machine Unlearning Verification

    cs.LG 2025-06 unverdicted novelty 6.0

    A survey that organizes machine unlearning verification methods into behavioral and parametric categories and outlines open problems.

  8. Not Every Subject Should Stay: Machine Unlearning for Noisy Engagement Recognition

    cs.CV 2026-05 unverdicted novelty 5.0

    Approximate subject-level unlearning recovers 89.3% and 92.5% of oracle performance gains on EngageNet and DAiSEE at roughly one-quarter the retraining cost in K=3 forget-set regimes.

  9. What Security and Privacy Transparency Users Need from Consumer-Facing Generative AI

    cs.HC 2026-04 unverdicted novelty 5.0

    A qualitative study of 21 GenAI users finds that current S&P transparency is often seen as incomplete or untrustworthy, leading to proxy-based adoption and constrained use, with calls for independent evaluations and o...

Reference graph

Works this paper leans on

163 extracted references · 163 canonical work pages · cited by 8 Pith papers · 2 internal anchors

  1. [1]

    Machine learning for internet of things data analysis: A survey

    Mohammad Saeid Mahdavinejad, Mohammadreza Rezvan, Mohammadamin Barekatain, Peyman Adibi, Payam Barnaghi, and Amit P Sheth. Machine learning for internet of things data analysis: A survey. Digital Communications and Networks , 4(3):161–175, 2018

    work page 2018

  2. [2]

    Overcoming catastrophic forgetting in neural networks

    James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, et al. Overcoming catastrophic forgetting in neural networks. Proceedings of the national academy of sciences , 114(13):3521–3526, 2017

    work page 2017

  3. [3]

    Experience replay for continual learning

    David Rolnick, Arun Ahuja, Jonathan Schwarz, Timothy Lillicrap, and Gregory Wayne. Experience replay for continual learning. Advances in neural information processing systems , 32, 2019

    work page 2019

  4. [4]

    A comprehensive survey on poisoning attacks and countermeasures in machine learning

    Zhiyi Tian, Lei Cui, Jie Liang, and Shui Yu. A comprehensive survey on poisoning attacks and countermeasures in machine learning. ACM Computing Surveys, 55(8):1–35, 2022

    work page 2022

  5. [5]

    The EU proposal for a general data protection regulation and the roots of the ’right to be forgotten’.Comput

    Alessandro Mantelero. The EU proposal for a general data protection regulation and the roots of the ’right to be forgotten’.Comput. Law Secur. Rev., 29(3):229–235, 2013

    work page 2013

  6. [6]

    of the Privacy Commissioner of Canada

    O. of the Privacy Commissioner of Canada. Announcement: Privacy commissioner seeks federal court determination on key issue for canadians’ online reputation

    work page

  7. [7]

    A guide to the california consumer privacy act of 2018

    Lydia de la Torre. A guide to the california consumer privacy act of 2018. A vailable at SSRN 3275571, 2018

    work page 2018

  8. [8]

    Towards making systems forget with machine unlearning

    Yinzhi Cao and Junfeng Yang. Towards making systems forget with machine unlearning. In 2015 IEEE Symposium on Security and Privacy, pages 463–480. IEEE, 2015

    work page 2015

  9. [9]

    Machine unlearning

    Lucas Bourtoule, Varun Chandrasekaran, Christopher A Choquette-Choo, Hengrui Jia, Adelin Travers, Baiwu Zhang, David Lie, and Nicolas Papernot. Machine unlearning. In 2021 IEEE Symposium on Security and Privacy (SP) , pages 141–159. IEEE, 2021

    work page 2021

  10. [10]

    Recent advances in convolutional neural networks

    Jiuxiang Gu, Zhenhua Wang, Jason Kuen, Lianyang Ma, Amir Shahroudy, Bing Shuai, Ting Liu, Xingxing Wang, Gang Wang, Jianfei Cai, et al. Recent advances in convolutional neural networks. Pattern recognition, 77:354–377, 2018

    work page 2018

  11. [11]

    Efficient processing of deep neural networks: A tutorial and survey

    Vivienne Sze, Yu-Hsin Chen, Tien-Ju Yang, and Joel S Emer. Efficient processing of deep neural networks: A tutorial and survey. Proceedings of the IEEE , 105(12):2295–2329, 2017

    work page 2017

  12. [12]

    Understanding black-box predictions via influence functions

    Pang Wei Koh and Percy Liang. Understanding black-box predictions via influence functions. In International conference on machine learning, pages 1885–1894. PMLR, 2017

    work page 2017

  13. [13]

    Variational bayesian unlearning

    Quoc Phong Nguyen, Bryan Kian Hsiang Low, and Patrick Jaillet. Variational bayesian unlearning. Advances in Neural Information Processing Systems, 33:16025–16036, 2020

    work page 2020

  14. [14]

    Mixed-privacy forgetting in deep networks

    Aditya Golatkar, Alessandro Achille, Avinash Ravichandran, Marzia Polito, and Stefano Soatto. Mixed-privacy forgetting in deep networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 792–801, 2021

    work page 2021

  15. [15]

    Amnesiac machine learning

    Laura Graves, Vineel Nagisetty, and Vijay Ganesh. Amnesiac machine learning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 11516–11524, 2021

    work page 2021

  16. [16]

    Formalizing data deletion in the context of the right to be forgotten

    Sanjam Garg, Shafi Goldwasser, and Prashant Nalini Vasudevan. Formalizing data deletion in the context of the right to be forgotten. In Anne Canteaut and Yuval Ishai, editors, Advances in Cryptology - EUROCRYPT 2020 - 39th Annual International Conference on the Theory and Applications of Cryptographic Techniques, Zagreb, Croatia, May 10-14, 2020, Proceedi...

    work page 2020

  17. [17]

    Machine unlearning for random forests

    Jonathan Brophy and Daniel Lowd. Machine unlearning for random forests. In International Conference on Machine Learning , pages 1092–1104. PMLR, 2021

    work page 2021

  18. [18]

    ARCANE: an efficient architecture for exact machine unlearning

    Haonan Yan, Xiaoguang Li, Ziyao Guo, Hui Li, Fenghua Li, and Xiaodong Lin. ARCANE: an efficient architecture for exact machine unlearning. In Luc De Raedt, editor, Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI 2022, Vienna, Austria, 23-29 July 2022 , pages 4006–4013. ijcai.org, 2022

    work page 2022

  19. [19]

    Recommendation unlearning

    Chong Chen, Fei Sun, Min Zhang, and Bolin Ding. Recommendation unlearning. In Proceedings of the ACM Web Conference 2022 , pages 2768–2777, 2022

    work page 2022

  20. [20]

    Forget me now: Fast and exact unlearning in neighborhood-based recommendation

    Sebastian Schelter, Mozhdeh Ariannezhad, and Maarten de Rijke. Forget me now: Fast and exact unlearning in neighborhood-based recommendation. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval , pages 2011–2015, 2023

    work page 2011

  21. [21]

    No matter how you slice it: Machine unlearning with sisa comes at the expense of minority classes

    Korbinian Koch and Marcus Soll. No matter how you slice it: Machine unlearning with sisa comes at the expense of minority classes. In 2023 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) , pages 622–637. IEEE, 2023

    work page 2023

  22. [22]

    Eraser: Machine unlearning in mlaas via an inference serving-aware approach

    Yuke Hu, Jian Lou, Jiaqi Liu, Feng Lin, Zhan Qin, and Kui Ren. Eraser: Machine unlearning in mlaas via an inference serving-aware approach. Proceedings of the 2024 ACM SIGSAC Conference on Computer and Communications Security , 2024

    work page 2024

  23. [23]

    Deltagrad: Rapid retraining of machine learning models

    Yinjun Wu, Edgar Dobriban, and Susan Davidson. Deltagrad: Rapid retraining of machine learning models. In International Conference on Machine Learning, pages 10355–10366. PMLR, 2020

    work page 2020

  24. [24]

    PUMA: performance unchanged model augmentation for training data removal

    Ga Wu, Masoud Hashemi, and Christopher Srinivasa. PUMA: performance unchanged model augmentation for training data removal. In Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 2022, Thirty-Fourth Conference on Innovative Applications of Artificial Intelligence, IAAI 2022, The Twelveth Symposium on Educational Advances in Artificial Intelligen...

    work page 2022

  25. [25]

    Hannun, and Laurens van der Maaten

    Chuan Guo, Tom Goldstein, Awni Y. Hannun, and Laurens van der Maaten. Certified data removal from machine learning models. In Proceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event , volume 119 of Proceedings of Machine Learning Research , pages 3832–3842. PMLR, 2020

    work page 2020

  26. [26]

    Remember what you want to forget: Algorithms for machine unlearning

    Ayush Sekhari, Jayadev Acharya, Gautam Kamath, and Ananda Theertha Suresh. Remember what you want to forget: Algorithms for machine unlearning. Advances in Neural Information Processing Systems , 34, 2021

    work page 2021

  27. [27]

    Descent-to-delete: Gradient-based methods for machine unlearning

    Seth Neel, Aaron Roth, and Saeed Sharifi-Malvajerdi. Descent-to-delete: Gradient-based methods for machine unlearning. InAlgorithmic Learning Theory, pages 931–962. PMLR, 2021

    work page 2021

  28. [28]

    Hedgecut: Maintaining randomised trees for low-latency machine unlearning

    Sebastian Schelter, Stefan Grafberger, and Ted Dunning. Hedgecut: Maintaining randomised trees for low-latency machine unlearning. In Proceedings of the 2021 International Conference on Management of Data , pages 1545–1557, 2021

    work page 2021

  29. [29]

    Making ai forget you: Data deletion in machine learning

    Antonio Ginart, Melody Guan, Gregory Valiant, and James Y Zou. Making ai forget you: Data deletion in machine learning. Advances in Neural Information Processing Systems , 32, 2019

    work page 2019

  30. [30]

    Eternal sunshine of the spotless net: Selective forgetting in deep networks

    Aditya Golatkar, Alessandro Achille, and Stefano Soatto. Eternal sunshine of the spotless net: Selective forgetting in deep networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 9304–9312, 2020

    work page 2020

  31. [31]

    Deep unlearning via randomized conditionally independent hessians

    Ronak Mehta, Sourav Pal, Vikas Singh, and Sathya N Ravi. Deep unlearning via randomized conditionally independent hessians. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 10422–10431, 2022

    work page 2022

  32. [32]

    Online forgetting process for linear regression models

    Yuantong Li, Chi-Hua Wang, and Guang Cheng. Online forgetting process for linear regression models. In Arindam Banerjee and Kenji Fukumizu, editors, The 24th International Conference on Artificial Intelligence and Statistics, AISTATS 2021, April 13-15, 2021, Virtual Event, volume 130 of Proceedings of Machine Learning Research , pages 217–225. PMLR, 2021....

    work page 2021

  33. [33]

    Machine unlearning of features and labels

    Alexander Warnecke, Lukas Pirch, Christian Wressnegger, and Konrad Rieck. Machine unlearning of features and labels. In 30th Annual Network and Distributed System Security Symposium, NDSS 2023, San Diego, California, USA, February 27 - March 3, 2023 . The Internet Society, 2023

    work page 2023

  34. [34]

    Markov chain monte carlo-based machine unlearning: Unlearning what needs to be forgotten

    Quoc Phong Nguyen, Ryutaro Oikawa, Dinil Mon Divakaran, Mun Choon Chan, and Bryan Kian Hsiang Low. Markov chain monte carlo-based machine unlearning: Unlearning what needs to be forgotten. In ASIA CCS ’22: ACM Asia Conference on Computer and Communications Security, Nagasaki, Japan, 30 May 2022 - 3 June 2022 , pages 351–363. ACM, 2022

    work page 2022

  35. [35]

    Knowledge removal in sampling-based bayesian inference

    Shaopeng Fu, Fengxiang He, and Dacheng Tao. Knowledge removal in sampling-based bayesian inference. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022 . OpenReview.net, 2022

    work page 2022

  36. [36]

    Knowledge-adaptation priors

    Mohammad Emtiyaz E Khan and Siddharth Swaroop. Knowledge-adaptation priors. Advances in Neural Information Processing Systems , 34:19757–19770, 2021

    work page 2021

  37. [37]

    Fast yet effective machine unlearning.IEEE Transactions on Neural Networks and Learning Systems , 2023

    Ayush K Tarun, Vikram S Chundawat, Murari Mandal, and Mohan Kankanhalli. Fast yet effective machine unlearning.IEEE Transactions on Neural Networks and Learning Systems , 2023

    work page 2023

  38. [38]

    Machine unlearning via representation forgetting with parameter self-sharing

    Weiqi Wang, Chenhan Zhang, Zhiyi Tian, and Shui Yu. Machine unlearning via representation forgetting with parameter self-sharing. IEEE Transactions on Information Forensics and Security , 2023

    work page 2023

  39. [39]

    Zero-shot machine unlearning

    Vikram S Chundawat, Ayush K Tarun, Murari Mandal, and Mohan Kankanhalli. Zero-shot machine unlearning. IEEE Transactions on Information Forensics and Security , 18:2345–2354, 2023

    work page 2023

  40. [40]

    Machine unlearning of features and labels

    Alexander Warnecke, Lukas Pirch, Christian Wressnegger, and Konrad Rieck. Machine unlearning of features and labels. 31th Annual Network and Distributed System Security Symposium, NDSS 2024 , 2024

    work page 2024

  41. [41]

    Muter: Machine unlearning on adversarially trained models

    Junxu Liu, Mingsheng Xue, Jian Lou, Xiaoyu Zhang, Li Xiong, and Zhan Qin. Muter: Machine unlearning on adversarially trained models. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 4892–4902, 2023

    work page 2023

  42. [42]

    Erm-ktp: Knowledge-level machine unlearning via knowledge transfer

    Shen Lin, Xiaoyu Zhang, Chenyang Chen, Xiaofeng Chen, and Willy Susilo. Erm-ktp: Knowledge-level machine unlearning via knowledge transfer. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 20147–20155, 2023

    work page 2023

  43. [43]

    Learning to unlearn: Instance-wise unlearning for pre-trained classifiers

    Sungmin Cha, Sungjun Cho, Dasol Hwang, Honglak Lee, Taesup Moon, and Moontae Lee. Learning to unlearn: Instance-wise unlearning for pre-trained classifiers. In Proceedings of the AAAI Conference on Artificial Intelligence , volume 38, pages 11186–11194, 2024

    work page 2024

  44. [44]

    Membership inference via backdooring

    Hongsheng Hu, Zoran Salcic, Gillian Dobbie, Jinjun Chen, Lichao Sun, and Xuyun Zhang. Membership inference via backdooring. In Luc De Raedt, editor, Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI 2022, Vienna, Austria, 23-29 July 2022, pages 3832–3838. ijcai.org, 2022

    work page 2022

  45. [45]

    Verifying in the dark: Verifiable machine unlearning by using invisible backdoor triggers

    Yu Guo, Yu Zhao, Saihui Hou, Cong Wang, and Xiaohua Jia. Verifying in the dark: Verifiable machine unlearning by using invisible backdoor triggers. IEEE Transactions on Information Forensics and Security , 2023

    work page 2023

  46. [46]

    Unrolling sgd: Understanding factors influencing machine unlearning

    Anvith Thudi, Gabriel Deza, Varun Chandrasekaran, and Nicolas Papernot. Unrolling sgd: Understanding factors influencing machine unlearning. In 2022 IEEE 7th European Symposium on Security and Privacy (EuroS&P) , pages 303–319. IEEE, 2022

    work page 2022

  47. [47]

    Learn to forget: Machine unlearning via neuron masking

    Zhuo Ma, Yang Liu, Ximeng Liu, Jian Liu, Jianfeng Ma, and Kui Ren. Learn to forget: Machine unlearning via neuron masking. IEEE Transactions on Dependable and Secure Computing , 2022

    work page 2022

  48. [48]

    Athena: Probabilistic verification of machine unlearning

    David Marco Sommer, Liwei Song, Sameer Wagh, and Prateek Mittal. Athena: Probabilistic verification of machine unlearning. Proceedings on Privacy Enhancing Technologies , 3:268–290, 2022. ACM Comput. Surv. Machine Unlearning: A Comprehensive Survey • 29

    work page 2022

  49. [49]

    EMA: auditing data removal from trained models

    Yangsibo Huang, Xiaoxiao Li, and Kai Li. EMA: auditing data removal from trained models. In Marleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, and Caroline Essert, editors, Medical Image Computing and Computer Assisted Intervention - MICCAI 2021 - 24th International Conference, Strasbourg, France, Septem...

    work page 2021

  50. [50]

    On the necessity of auditable algorithmic definitions for machine unlearning

    Anvith Thudi, Hengrui Jia, Ilia Shumailov, and Nicolas Papernot. On the necessity of auditable algorithmic definitions for machine unlearning. In 31st USENIX Security Symposium (USENIX Security 22) , pages 4007–4022, 2022

    work page 2022

  51. [51]

    Certified graph unlearning

    Eli Chien, Chao Pan, and Olgica Milenkovic. Certified graph unlearning. arXiv preprint arXiv:2206.09140, 2022

    work page arXiv 2022

  52. [52]

    Graph unlearning

    Min Chen, Zhikun Zhang, Tianhao Wang, Michael Backes, Mathias Humbert, and Yang Zhang. Graph unlearning. In Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security , pages 499–513, 2022

    work page 2022

  53. [53]

    Towards effective and general graph unlearning via mutual evolution

    Xunkai Li, Yulin Zhao, Zhengyu Wu, Wentao Zhang, Rong-Hua Li, and Guoren Wang. Towards effective and general graph unlearning via mutual evolution. In Proceedings of the AAAI Conference on Artificial Intelligence , volume 38, pages 13682–13690, 2024

    work page 2024

  54. [54]

    Privacy matters! efficient graph representation unlearning with data removal guarantee

    Weilin Cong and Mehrdad Mahdavi. Privacy matters! efficient graph representation unlearning with data removal guarantee. 2022

    work page 2022

  55. [55]

    Inductive graph unlearning

    Cheng-Long Wang, Mengdi Huai, and Di Wang. Inductive graph unlearning. In 32nd USENIX Security Symposium (USENIX Security 23), pages 3205–3222, 2023

    work page 2023

  56. [56]

    Certified edge unlearning for graph neural networks

    Kun Wu, Jie Shen, Yue Ning, Ting Wang, and Wendy Hui Wang. Certified edge unlearning for graph neural networks. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , pages 2606–2617, 2023

    work page 2023

  57. [57]

    Forgetting and remembering are both you need: Balanced graph structure unlearning

    Chenhan Zhang, Weiqi Wang, Zhiyi Tian, and Shui Yu. Forgetting and remembering are both you need: Balanced graph structure unlearning. IEEE Transactions on Information Forensics and Security , 2024

    work page 2024

  58. [58]

    Gif: A general graph unlearning strategy via influence function

    Jiancan Wu, Yi Yang, Yuchun Qian, Yongduo Sui, Xiang Wang, and Xiangnan He. Gif: A general graph unlearning strategy via influence function. In Proceedings of the ACM Web Conference 2023 , pages 651–661, 2023

    work page 2023

  59. [59]

    Federated unlearning via class-discriminative pruning

    Junxiao Wang, Song Guo, Xin Xie, and Heng Qi. Federated unlearning via class-discriminative pruning. In Proceedings of the ACM Web Conference 2022, pages 622–632, 2022

    work page 2022

  60. [60]

    How to forget clients in federated online learning to rank? In European Conference on Information Retrieval, pages 105–121

    Shuyi Wang, Bing Liu, and Guido Zuccon. How to forget clients in federated online learning to rank? In European Conference on Information Retrieval, pages 105–121. Springer, 2024

    work page 2024

  61. [61]

    Sifu: Sequential informed federated unlearning for efficient and provable client unlearning in federated optimization

    Yann Fraboni, Martin Van Waerebeke, Kevin Scaman, Richard Vidal, Laetitia Kameni, and Marco Lorenzi. Sifu: Sequential informed federated unlearning for efficient and provable client unlearning in federated optimization. In International Conference on Artificial Intelligence and Statistics, pages 3457–3465. PMLR, 2024

    work page 2024

  62. [62]

    Bfu: Bayesian federated unlearning with parameter self-sharing

    Weiqi Wang, Zhiyi Tian, Chenhan Zhang, An Liu, and Shui Yu. Bfu: Bayesian federated unlearning with parameter self-sharing. In Proceedings of the 2023 ACM Asia Conference on Computer and Communications Security , pages 567–578, 2023

    work page 2023

  63. [63]

    Federaser: Enabling efficient client-level data removal from federated learning models

    Gaoyang Liu, Xiaoqiang Ma, Yang Yang, Chen Wang, and Jiangchuan Liu. Federaser: Enabling efficient client-level data removal from federated learning models. In 2021 IEEE/ACM 29th International Symposium on Quality of Service (IWQOS) , pages 1–10. IEEE, 2021

    work page 2021

  64. [64]

    The right to be forgotten in federated learning: An efficient realization with rapid retraining

    Yi Liu, Lei Xu, Xingliang Yuan, Cong Wang, and Bo Li. The right to be forgotten in federated learning: An efficient realization with rapid retraining. In IEEE INFOCOM 2022 - IEEE Conference on Computer Communications, London, United Kingdom, May 2-5, 2022 , pages 1749–1758. IEEE, 2022

    work page 2022

  65. [65]

    Fedrecovery: Differentially private machine unlearning for federated learning frameworks

    Lefeng Zhang, Tianqing Zhu, Haibin Zhang, Ping Xiong, and Wanlei Zhou. Fedrecovery: Differentially private machine unlearning for federated learning frameworks. IEEE Transactions on Information Forensics and Security , 2023

    work page 2023

  66. [66]

    Adaptive machine unlearning

    Varun Gupta, Christopher Jung, Seth Neel, Aaron Roth, Saeed Sharifi-Malvajerdi, and Chris Waites. Adaptive machine unlearning. Advances in Neural Information Processing Systems , 34:16319–16330, 2021

    work page 2021

  67. [67]

    Federated unlearning with knowledge distillation

    Chen Wu, Sencun Zhu, and Prasenjit Mitra. Federated unlearning with knowledge distillation. arXiv preprint arXiv:2201.09441, 2022

    work page arXiv 2022

  68. [68]

    Data unlearning in diffusion models.arXiv preprint arXiv:2503.01034, 2025

    Silas Alberti, Kenan Hasanaliyev, Manav Shah, and Stefano Ermon. Data unlearning in diffusion models.arXiv preprint arXiv:2503.01034, 2025

    work page arXiv 2025

  69. [69]

    Score forgetting distillation: A swift, data-free method for machine unlearning in diffusion models

    Tianqi Chen, Shujian Zhang, and Mingyuan Zhou. Score forgetting distillation: A swift, data-free method for machine unlearning in diffusion models. arXiv preprint arXiv:2409.11219, 2024

    work page arXiv 2024

  70. [70]

    Erasing undesirable influence in diffusion models

    Jing Wu, Trung Le, Munawar Hayat, and Mehrtash Harandi. Erasing undesirable influence in diffusion models. In Proceedings of the Computer Vision and Pattern Recognition Conference , pages 28263–28273, 2025

    work page 2025

  71. [71]

    Machine unlearning for image-to-image generative models

    Guihong Li, Hsiang Hsu, Chun-Fu Chen, and Radu Marculescu. Machine unlearning for image-to-image generative models. arXiv preprint arXiv:2402.00351, 2024

    work page arXiv 2024

  72. [72]

    Defensive unlearning with adversarial training for robust concept erasure in diffusion models

    Yimeng Zhang, Xin Chen, Jinghan Jia, Yihua Zhang, Chongyu Fan, Jiancheng Liu, Mingyi Hong, Ke Ding, and Sijia Liu. Defensive unlearning with adversarial training for robust concept erasure in diffusion models. Advances in neural information processing systems , 37:36748–36776, 2024

    work page 2024

  73. [73]

    Unlearning concepts in diffusion model via concept domain correction and concept preserving gradient

    Yongliang Wu, Shiji Zhou, Mingzhuo Yang, Lianzhe Wang, Heng Chang, Wenbo Zhu, Xinting Hu, Xiao Zhou, and Xu Yang. Unlearning concepts in diffusion model via concept domain correction and concept preserving gradient. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 8496–8504, 2025

    work page 2025

  74. [74]

    Saeuron: Interpretable concept unlearning in diffusion models with sparse autoencoders

    Bartosz Cywiński and Kamil Deja. Saeuron: Interpretable concept unlearning in diffusion models with sparse autoencoders. arXiv preprint arXiv:2501.18052, 2025. ACM Comput. Surv. 30 • Weiqi Wang et al

    work page arXiv 2025

  75. [75]

    Unlearncanvas: Stylized image dataset for enhanced machine unlearning evaluation in diffusion models

    Yihua Zhang, Chongyu Fan, Yimeng Zhang, Yuguang Yao, Jinghan Jia, Jiancheng Liu, Gaoyuan Zhang, Gaowen Liu, Ramana Rao Kompella, Xiaoming Liu, et al. Unlearncanvas: Stylized image dataset for enhanced machine unlearning evaluation in diffusion models. arXiv preprint arXiv:2402.11846, 2024

    work page arXiv 2024

  76. [76]

    Boosting alignment for post-unlearning text-to-image generative models

    Myeongseob Ko, Henry Li, Zhun Wang, Jonathan Patsenker, Jiachen Tianhao Wang, Qinbin Li, Ming Jin, Dawn Song, and Ruoxi Jia. Boosting alignment for post-unlearning text-to-image generative models. Advances in Neural Information Processing Systems , 37: 85131–85154, 2024

    work page 2024

  77. [77]

    The illusion of unlearning: The unstable nature of machine unlearning in text-to-image diffusion models

    Naveen George, Karthik Nandan Dasaraju, Rutheesh Reddy Chittepu, and Konda Reddy Mopuri. The illusion of unlearning: The unstable nature of machine unlearning in text-to-image diffusion models. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 13393–13402, 2025

    work page 2025

  78. [78]

    Large language model unlearning

    Yuanshun Yao, Xiaojun Xu, and Yang Liu. Large language model unlearning. Advances in Neural Information Processing Systems , 37: 105425–105475, 2024

    work page 2024

  79. [79]

    Machine unlearning of pre-trained large language models

    Jin Yao, Eli Chien, Minxin Du, Xinyao Niu, Tianhao Wang, Zezhou Cheng, and Xiang Yue. Machine unlearning of pre-trained large language models. arXiv preprint arXiv:2402.15159, 2024

    work page arXiv 2024

  80. [80]

    Rethinking machine unlearning for large language models

    Sijia Liu, Yuanshun Yao, Jinghan Jia, Stephen Casper, Nathalie Baracaldo, Peter Hase, Yuguang Yao, Chris Yuhao Liu, Xiaojun Xu, Hang Li, et al. Rethinking machine unlearning for large language models. Nature Machine Intelligence, pages 1–14, 2025

    work page 2025

Showing first 80 references.