ATWU jointly optimizes model parameters and token weights via a linear scorer on hidden states, recovering oracle forget-specific tokens under a separation condition and achieving SOTA forget-retain trade-offs on TOFU and RWKU.
hub
Minseok Choi, Kyunghyun Min, and Jaegul Choo
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
SAEParate disentangles sparse representations in diffusion models via contrastive clustering and nonlinear encoding to enable more precise concept unlearning with reduced side effects.
MemoRepair formalizes the cascade update problem in agentic memory and solves it via a min-cut reduction that eliminates invalidated memory exposure to 0% while recovering 91-94% of valid successors at 57-76% of baseline repair cost.
LLM unlearning is reframed as inadvertently installing backdoor triggers on forget-tokens; Random Noise Augmentation is introduced as a defense that improves robustness with theoretical guarantees.
Unlearning in multilingual LLMs suppresses rather than erases knowledge in later layers, with transfer varying by language similarity and reversible via inference-time steering.
Interpretability research should be judged by actionability—the degree to which its insights support concrete decisions and interventions—rather than explanatory power alone.
Adaptive Unlearning suppresses package hallucinations in code-generating LLMs by 81% while preserving benchmark performance, using model-generated data and no human labels.
Harmful generation in LLMs relies on a compact, unified set of weights that alignment compresses and that are distinct from benign capabilities, explaining emergent misalignment.
Survey identifying technical and supply-chain barriers to GDPR data subject rights in ML, with new framing of 'models in the dark' for downstream opacity.
Unlearned language models retain low calibration error but show increased shortcut reliance on the TOFU benchmark, extending the reliability paradox to machine unlearning.
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.
citing papers explorer
-
Disentangled Sparse Representations for Concept-Separated Diffusion Unlearning
SAEParate disentangles sparse representations in diffusion models via contrastive clustering and nonlinear encoding to enable more precise concept unlearning with reduced side effects.
-
LLM Ghostbusters: Surgical Hallucination Suppression via Adaptive Unlearning
Adaptive Unlearning suppresses package hallucinations in code-generating LLMs by 81% while preserving benchmark performance, using model-generated data and no human labels.