machine-unlearning

The paper identifies sequential reappearance as a failure mode in diffusion data-point unlearning and proposes a sharpness-guided method to improve forgetting persistence across deletion sequences.

The paper introduces GUARD, a method for natural forgetting in large reasoning models that uses guided answer-reasoning distillation to suppress unsafe or private content in chain-of-thought traces while preserving reasoning utility.

This paper introduces μ²-Bench, a benchmark for evaluating multilingual machine unlearning in large language models, aiming to ensure that undesired information is effectively removed across diverse languages.

ARIA is a test-time unlearning method for large language models that uses sparse autoencoders to suppress unwanted knowledge during inference without modifying weights, improving the forget-retain trade-off and remaining robust to adversarial attacks.

This paper explores machine unlearning techniques for Large Audio-Language Models to remove sensitive information from speech QA tasks, demonstrating methods that reduce privacy leakage by up to 80% while maintaining performance.

The paper introduces forget-set misalignment in LLM unlearning and proposes a data-blind framework called CONFS to address it, achieving a competitive forgetting-utility balance.

This paper introduces I-CARE, a methodology for systematically analyzing interference in machine unlearning for text-to-image models, providing formal definitions and an open-source framework to enable reproducible study.

This paper proposes CallosumNet, a biologically inspired framework for efficient unlearning in spatio-temporal graphs to comply with privacy regulations like GDPR, achieving complete unlearning with minimal accuracy loss.

IsleNet introduces a spatial-entropy-based partitioning method for spatiotemporal graph unlearning, enabling exact data removal with low computational cost while maintaining high accuracy for privacy regulations.

The study reveals substantial gaps in machine unlearning for LLMs, showing that adversarial evaluation uncovers recoverability of forgotten information despite strong standard metrics, highlighting the need for adversarial stress-testing.

This paper identifies tool-mediated recovery as a failure mode in LLM unlearning and proposes Agentic Tool Unlearning (ATU) to reduce both parametric recall and tool-based recovery while preserving normal tool use.

The paper introduces ConceptGuard, a benchmark for evaluating context-sensitive unlearning in large language models using dual-use concepts, revealing that current unlearning techniques perform poorly under this practical evaluation framework.

This paper explores forget set curation for machine unlearning in language models, introducing a benchmark to evaluate verbatim output suppression and highlighting trade-offs between effectiveness and capability retention.

The paper proposes AdaPop, an adaptive popularity-based method for LLM unlearning that adjusts gradient pressure based on fact frequency to improve forgetting effectiveness and reduce leakage under queries.

This paper identifies a privacy vulnerability in RL-trained multimodal large reasoning models, which can leak sensitive facts in their reasoning traces even after unlearning, and proposes LEMUR, a training-free inference-time framework that uses entropy dynamics to detect and suppress such leakage.

Proposes J-Access, an inference-time audit using the Jacobian lens to measure residual knowledge accessibility in unlearned LLMs, finding that accessibility predicts recovery speed but that directly minimizing it fails to promote genuine deletion.

Introduces GROM, a gradient-free one-shot machine unlearning method that computes a closed-form additive weight update via ridge-regularized least squares, achieving state-of-the-art forgetting-utility trade-offs on benchmarks like TOFU and WMDP, and resisting quantization-based recovery attacks.

This paper introduces a new benchmark for evaluating machine unlearning, focusing on multi-hop reasoning consistency and recovery robustness. Experiments show existing unlearning methods face an 'impossible triangle' trade-off among forget quality, robustness, and utility preservation.

This paper proposes a retain-aware localization method for machine unlearning that reduces collateral damage to semantically similar retained examples, and introduces a retain-similar evaluation set. Experiments on CIFAR-10 with ResNet18 show reduced collateral damage and improved unlearning metrics.

This paper investigates exact deletion from language-model memory, showing that subtractive methods work when record influence is addressable, while replay/rebuild is needed when influence is woven into recurrent state. Experiments on Gemma and Kimi hybrid models demonstrate trade-offs in utility and exactness.