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.
Machine Unlearning: A Comprehensive Survey
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
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.
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
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.
VGID constructs an intervention-induced teacher distribution via visual perturbation plus textual in-context unlearning and distills it into the student MLLM to achieve parameter-level forgetting.
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.
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.
PeCL applies token-level dynamic differential privacy and privacy-guided memory sculpting to achieve superior privacy-utility balance in continual learning.
A survey that organizes machine unlearning verification methods into behavioral and parametric categories and outlines open problems.
Introduces Grouped Memorization Evaluation and FedMemPrune to remove unique memorized information in federated unlearning while preserving overlapping knowledge.
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.
DiRLU distills an A2C teacher into a lightweight student that detects BoT-IoT attacks at 99.6% accuracy with 2370 FLOPS and reversible post-hoc feature unlearning.
A qualitative study of 21 U.S. GenAI users reveals that existing security and privacy transparency is perceived as ineffective and lacking credibility, leading users to rely on proxies like popularity and constraining high-stakes use.
citing papers explorer
-
Knowledge Beyond Language: Bridging the Gap in Multilingual Machine Unlearning Evaluation
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.
-
Copyright Laundering Through the AI Ouroboros: Adapting the 'Fruit of the Poisonous Tree' Doctrine to Recursive AI Training
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.
-
Visual-Noise Guided In-Context Distillation for Multimodal Large Language Model Unlearning
VGID constructs an intervention-induced teacher distribution via visual perturbation plus textual in-context unlearning and distills it into the student MLLM to achieve parameter-level forgetting.
-
Class Unlearning via Depth-Aware Removal of Forget-Specific Directions
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.
-
WIN-U: Woodbury-Informed Newton-Unlearning as a retain-free Machine Unlearning Framework
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.
-
Forget What's Sensitive, Remember What Matters: Token-Level Differential Privacy in Memory Sculpting for Continual Learning
PeCL applies token-level dynamic differential privacy and privacy-guided memory sculpting to achieve superior privacy-utility balance in continual learning.
-
Towards Reliable Forgetting: A Survey on Machine Unlearning Verification
A survey that organizes machine unlearning verification methods into behavioral and parametric categories and outlines open problems.
-
Rethinking Federated Unlearning via the Lens of Memorization
Introduces Grouped Memorization Evaluation and FedMemPrune to remove unique memorized information in federated unlearning while preserving overlapping knowledge.
-
Not Every Subject Should Stay: Machine Unlearning for Noisy Engagement Recognition
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.
-
Unlearning to Protect: A Distilled Reinforcement Learning Framework with Privacy-Preserving Feature Unlearning and XAI for IoT Security
DiRLU distills an A2C teacher into a lightweight student that detects BoT-IoT attacks at 99.6% accuracy with 2370 FLOPS and reversible post-hoc feature unlearning.
-
Understanding U.S. Users' Security and Privacy Transparency Needs for Consumer-Facing Generative AI
A qualitative study of 21 U.S. GenAI users reveals that existing security and privacy transparency is perceived as ineffective and lacking credibility, leading users to rely on proxies like popularity and constraining high-stakes use.