Forgetting Only What Matters: Layer-Selective Unlearning toward Robust LLMs

• URL: http://arxiv.org/abs/2609.10439v1
• Date: Wed, 09 Sep 2026 16:50:41 GMT
• Title: Forgetting Only What Matters: Layer-Selective Unlearning toward Robust LLMs
• Abstract summary: Large Language Models (LLMs) can memorize and reproduce sensitive, copyrighted, or otherwise undesirable training content.<n>We propose Forgetting Only What Matters via Unlearning Layers (FOM-UL), a layer-level unlearning framework that selects transformer layers using a forget-to-retain significance score.
• Score: 1.988145627448243
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) can memorize and reproduce sensitive, copyrighted, or otherwise undesirable training content, creating privacy, safety, and regulatory concerns. Machine unlearning offers a practical alternative to full retraining, but many existing methods apply broad or fixed parameter updates that can degrade utility and remain brittle under deployment changes such as post-training quantization, where forgotten knowledge may partially re-emerge. We propose Forgetting Only What Matters via Unlearning Layers (FOM-UL), a layer-level unlearning framework that selects transformer layers using a forget-to-retain significance score. This score identifies layers with high influence on the forget set and low sensitivity to the retain set, allowing FOM-UL to concentrate updates where they are most effective while leaving most of the model unchanged. This targeted update strategy improves the forgetting-utility trade-off and provides an empirical path toward quantization-resilient unlearning by reducing the chance that small, diffuse updates are erased by low-bit rounding. Across TOFU, KnowUnDo, and MUSE-style evaluations, FOM-UL reduces residual memorization compared with strong GA, NPO, KLD, SURE, ReLearn, and LUNAR-based baselines while preserving retain-set utility close to the vanilla model. Under 8-bit and 4-bit post-training quantization, FOM-UL maintains stronger memorization suppression and utility preservation than competing methods, and adversarial prompt evaluations show lower recovery of forgotten content. Overall, FOM-UL provides an efficient unlearning strategy that improves targeted forgetting, utility preservation, and deployment robustness without claiming formal guarantees of erasure.

Related papers

• Test-Time Unlearning via Sparse Autoencoder [55.4290003355764]
  We propose ARIA, a test-time unlearning method that leaves model weights intact and gates access to unwanted knowledge only when generation enters a forget-related state.<n>ARIA improves the forget-retain trade-off over weight-based baselines across both a thinking model and an instruction model.
  arXiv  Detail & Related papers  (2026-09-14T18:58:09Z)
• Scaling Adaptive Depth with Norm-Agnostic Residual Networks [9.219329760729629]
  NAG is a norm-agnostic residual architecture that separates magnitude from directional information in the residual stream.<n>NAG introduces only a negligible number of additional parameters and relies on simple operations that are easily kernel-fusible.<n>We show that this architecture outperforms baseline Transformers, with gains that increase substantially as depth grows.
  arXiv  Detail & Related papers  (2026-06-15T02:05:37Z)
• MIITA: Memory-Induced Inference-Time Adaptation for Continual Learning with Small Language Models [64.2820121997882]
  MIITA is a Memory-Induced Inference-Time Adaptation framework for supervised CL under constrained storage.<n>It stores supervised experiences as compact correction-direction prototypes with semantic anchors, and retrieves them at inference time using semantic and uncertainty-based cues.<n>Experiments show that MIITA consistently improves final performance and mitigates forgetting under fixed memory budgets.
  arXiv  Detail & Related papers  (2026-05-20T03:03:04Z)
• Less is More: Geometric Unlearning for LLMs with Minimal Data Disclosure [10.0091317316944]
  Geometric Unlearning (GU) operates directly on the model's prompt-time planning states without access to the original training corpus.<n> GU distills a compact, low-rank geometry of desired safe behavior from a small set of safe reference prompts.<n>A teacher-distillation regularizer on synthetic non-target anchors further reduces collateral drift.
  arXiv  Detail & Related papers  (2026-05-03T06:20:03Z)
• FIT: Defying Catastrophic Forgetting in Continual LLM Unlearning [23.471857001200465]
  fit is a framework for continual unlearning that handles large numbers of deletion requests.<n>fit mitigates degradation through rigorous data underlineFiltering, underlineImportance-aware updates, and underlineTargeted layer attribution.<n>fit surpasses existing methods on MMLU, CommonsenseQA, and GSM8K, and remains resistant against both relearning and quantization recovery attacks.
  arXiv  Detail & Related papers  (2026-01-29T13:15:32Z)
• Downgrade to Upgrade: Optimizer Simplification Enhances Robustness in LLM Unlearning [25.53799024782883]
  Large language model (LLM) unlearning aims to surgically remove the influence of undesired data or knowledge from an existing model.<n>Recent findings reveal that unlearning manipulations such as weight quantization or fine-tuning can quickly neutralize the intended forgetting.
  arXiv  Detail & Related papers  (2025-10-01T10:50:14Z)
• Unlearning That Lasts: Utility-Preserving, Robust, and Almost Irreversible Forgetting in LLMs [31.768387661474904]
  Unlearning in large language models (LLMs) involves precisely removing specific information from a pre-trained model.<n>This is crucial to ensure safety of LLMs by deleting private data or harmful knowledge acquired during pre-training.<n>We introduce JensUn, where we leverage the Jensen-Shannon Divergence as the training objective for both forget and retain sets.<n>In extensive experiments, JensUn achieves better forget-utility trade-off than competing methods, and even demonstrates strong resilience to benign relearning.
  arXiv  Detail & Related papers  (2025-09-02T20:38:53Z)
• GUARD: Guided Unlearning and Retention via Data Attribution for Large Language Models [17.83305806604326]
  GUARD is a framework for guided unlearning and retention via data attribution.<n>It assigns adaptive, nonuniform unlearning weights to samples, inversely proportional to their proxy attribution scores.<n>We provide rigorous theoretical guarantees that GUARD significantly improves retention while maintaining forgetting metrics comparable to prior methods.
  arXiv  Detail & Related papers  (2025-06-12T17:49:09Z)
• BLUR: A Bi-Level Optimization Approach for LLM Unlearning [100.90394814817965]
  We argue that it is important to model the hierarchical structure of the unlearning problem.<n>We propose a novel algorithm, termed Bi-Level UnleaRning (textttBLUR), which delivers superior performance.
  arXiv  Detail & Related papers  (2025-06-09T19:23:05Z)
• UniErase: Towards Balanced and Precise Unlearning in Language Models [69.04923022755547]
  Large language models (LLMs) require iterative updates to address the outdated information problem.<n>UniErase is a novel unlearning framework that demonstrates precision and balanced performances between knowledge unlearning and ability retaining.
  arXiv  Detail & Related papers  (2025-05-21T15:53:28Z)
• ReLearn: Unlearning via Learning for Large Language Models [64.2802606302194]
  We propose ReLearn, a data augmentation and fine-tuning pipeline for effective unlearning.<n>This framework introduces Knowledge Forgetting Rate (KFR) and Knowledge Retention Rate (KRR) to measure knowledge-level preservation.<n>Our experiments show that ReLearn successfully achieves targeted forgetting while preserving high-quality output.
  arXiv  Detail & Related papers  (2025-02-16T16:31:00Z)
• Continuous Knowledge-Preserving Decomposition with Adaptive Layer Selection for Few-Shot Class-Incremental Learning [73.59672160329296]
  CKPD-FSCIL is a unified framework that unlocks the underutilized capacity of pretrained weights.<n>Our method consistently outperforms state-of-the-art approaches in both adaptability and knowledge retention.
  arXiv  Detail & Related papers  (2025-01-09T07:18:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.