Pith. sign in

REVIEW 13 cited by

Unlearn What You Want to Forget: Efficient Unlearning for LLMs

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 2310.20150 v1 pith:CVS32SIJ submitted 2023-10-31 cs.CL cs.AI

Unlearn What You Want to Forget: Efficient Unlearning for LLMs

classification cs.CL cs.AI
keywords dataunlearningllmsdifferentefficientforgetissueslayers
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Large language models (LLMs) have achieved significant progress from pre-training on and memorizing a wide range of textual data, however, this process might suffer from privacy issues and violations of data protection regulations. As a result, the ability to easily remove data related to individual users from such models while not deteriorating their predictive quality after the removal becomes increasingly important. To address these issues, in this work, we propose an efficient unlearning framework that could efficiently update LLMs without having to retrain the whole model after data removals, by introducing lightweight unlearning layers learned with a selective teacher-student objective into the transformers. In addition, we introduce a fusion mechanism to effectively combine different unlearning layers that learns to forget different sets of data to handle a sequence of forgetting operations. Experiments on classification and generation tasks demonstrate the effectiveness of our proposed methods compared to the state-of-the-art baselines.

discussion (0)

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

Forward citations

Cited by 13 Pith papers

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

  1. DurableUn: Quantization-Induced Recovery Attacks in Machine Unlearning

    cs.LG 2026-05 conditional novelty 8.0

    INT4 quantization recovers up to 22 times more forgotten training data in unlearned LLMs, and the proposed DURABLEUN-SAF method is the first to maintain forgetting across BF16, INT8, and INT4 precisions.

  2. Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning

    cs.LG 2024-04 conditional novelty 8.0

    NPO enables stable unlearning of 50%+ training data in LLMs on TOFU by making collapse exponentially slower than gradient ascent, preserving sensible outputs where prior methods fail.

  3. DurableUn: Quantization-Induced Recovery Attacks in Machine Unlearning

    cs.LG 2026-05 unverdicted novelty 7.0

    INT4 quantization recovers forgotten data in unlearned LLMs up to 22x, exposing a trilemma with no existing method solving forgetting, utility, and robustness together; a new sharpness-aware method achieves cross-prec...

  4. Unlearning What Matters: Token-Level Attribution for Precise Language Model Unlearning

    cs.CL 2026-05 unverdicted novelty 6.0

    TokenUnlearn identifies critical tokens via masking and entropy signals then applies hard selection or soft weighting to unlearn only those tokens, yielding better forgetting and retained utility than sequence-level b...

  5. Representation-Guided Parameter-Efficient LLM Unlearning

    cs.CL 2026-04 unverdicted novelty 6.0

    REGLU guides LoRA-based unlearning via representation subspaces and orthogonal regularization to outperform prior methods on forget-retain trade-off in LLM benchmarks.

  6. Maximizing Local Entropy Where It Matters: Prefix-Aware Localized LLM Unlearning

    cs.CL 2026-01 conditional novelty 6.0

    PALU improves LLM unlearning by restricting entropy maximization to sensitive prefixes and top-k logits, achieving better forgetting with less utility loss.

  7. Maximizing Local Entropy Where It Matters: Prefix-Aware Localized LLM Unlearning

    cs.CL 2026-01 conditional novelty 6.0

    PALU shows that unlearning only needs local intervention—the first few tokens of the sensitive span and the top-k logits—not full-sequence, full-vocabulary suppression.

  8. Distribution Preference Optimization: A Fine-grained Perspective for LLM Unlearning

    cs.LG 2025-10 conditional novelty 6.0

    DiPO is a distribution-level unlearning method that constructs preference distributions from the model's own high-confidence logits and achieves state-of-the-art forget quality on TOFU while preserving utility.

  9. OFMU: Optimization-Driven Framework for Machine Unlearning

    cs.LG 2025-09 unverdicted novelty 6.0

    A penalty-based bi-level optimization framework for machine unlearning that decorrelates forget and retention gradients via inner maximization and restores utility via outer minimization, with convergence guarantees a...

  10. Towards Scalable Lifelong Knowledge Editing with Selective Knowledge Suppression

    cs.AI 2026-04 unverdicted novelty 5.0

    LightEdit enables scalable lifelong knowledge editing in LLMs via selective knowledge retrieval and probability suppression during decoding, outperforming prior methods on ZSRE, Counterfact, and RIPE while reducing tr...

  11. Forgetting to Witness: Efficient Federated Unlearning and Its Visible Evaluation

    cs.LG 2026-04 unverdicted novelty 5.0

    A complete pipeline for federated unlearning via knowledge distillation for efficient removal and a GAN-integrated classifier for visual evaluation of forgetting capacity.

  12. Revisiting the Past: Data Unlearning with Model State History

    cs.LG 2025-06 unverdicted novelty 5.0

    MSA performs data unlearning in LLMs by arithmetic operations on prior model checkpoints to remove targeted datapoint influence, with experiments showing competitive or better results than existing unlearning methods.

  13. AI as a Tool for Simulation-Based Experiments in Literary Studies

    cs.CL 2026-06 unverdicted novelty 4.0

    Proposes AI-driven simulations for literary-historical experiments and reports preliminary text-generation results claiming the first limited in-distribution outputs matching human novels.