Pith. sign in

REVIEW 9 cited by

A Comprehensive Survey of Machine Unlearning Techniques for Large Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.01854 v2 pith:TAIX6MRP submitted 2025-02-22 cs.CL cs.AI

classification cs.CLcs.AI
keywords unlearningcomprehensivecurrentexistinglanguagelargellmsmachine
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

This study investigates the machine unlearning techniques within the context of large language models (LLMs), referred to as \textit{LLM unlearning}. LLM unlearning offers a principled approach to removing the influence of undesirable data (e.g., sensitive or illegal information) from LLMs, while preserving their overall utility without requiring full retraining. Despite growing research interest, there is no comprehensive survey that systematically organizes existing work and distills key insights; here, we aim to bridge this gap. We begin by introducing the definition and the paradigms of LLM unlearning, followed by a comprehensive taxonomy of existing unlearning studies. Next, we categorize current unlearning approaches, summarizing their strengths and limitations. Additionally, we review evaluation metrics and benchmarks, providing a structured overview of current assessment methodologies. Finally, we outline promising directions for future research, highlighting key challenges and opportunities in the field.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 9 Pith papers

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

  1. Gradient Concentration, Not Weight Saliency, Explains Representation-Level Class Unlearning

    cs.LG 2026-07 accept novelty 7.0 of 10

    In class unlearning on CIFAR-10/100 with ResNet-18, the identity of saliency-selected weights does not affect representation-level recovery; late-layer gradient concentration and representation geometry drive the outcome.

  2. One Modality to Forget Them All: Enhancing Cross-Modal Unlearning in Vision-Language Models

    cs.CV 2026-07 conditional novelty 6.5 of 10

    Cross-modal unlearning transfer in vision-language models is asymmetric, architecture-dependent, and shallow under typographic attacks; influence-guided block selection reduces the measured gap.

  3. Anatomy of Unlearning: The Dual Impact of Fact Salience and Model Fine-Tuning

    cs.CL 2026-02 conditional novelty 6.0 of 10

    SFT models forget facts more stably than pretrained models, with 10-50% higher retention of unrelated knowledge when using the DUET benchmark of 28.6k Wikidata triplets.

  4. LLM Unlearning Should Be Form-Independent

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Existing LLM unlearning is form-dependent; the new ORT benchmark measures this, and the training-free ROCR edit reduces it by redirecting concept representations.

  5. Model Unlearning via Sparse Autoencoder Subspace Guided Projections

    cs.CL 2025-05 conditional novelty 6.0 of 10

    SSPU uses SAE-derived subspaces to guide weight updates, lowering WMDP-Cyber accuracy by 3.22% more than RMU while largely preserving MMLU, TruthfulQA, and GSM8K performance.

  6. Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks

    cs.LG 2026-07 conditional novelty 5.0 of 10

    A system-first taxonomy and literature synthesis of multimodal unlearning across vision, language, video, and audio, with datasets, benchmarks, metrics, applications, and open challenges.

  7. BalDRO: A Distributionally Robust Optimization based Framework for Large Language Model Unlearning

    cs.LG 2026-01 conditional novelty 5.0 of 10

    BalDRO makes LLM unlearning more balanced by updating against a worst-case-weighted forget distribution, improving forget quality on TOFU/MUSE at stable utility.

  8. VSCBench: Bridging the Gap in Vision-Language Model Safety Calibration

    cs.IR 2025-05 conditional novelty 5.0 of 10

    A new benchmark, VSCBench, measures oversafety and undersafety in vision-language models and shows that most models, including proprietary ones, are miscalibrated on at least one safety dimension.

  9. Get Experience from Practice: LLM Agents with Record & Replay

    cs.LG 2025-05 reject novelty 4.0 of 10

    AgentRR is a proposed paradigm that records agent traces, generalizes them into multi-level experiences, and replays them under safety checks to make LLM agents cheaper, faster, and more reliable.

Pith tools