CURE disentangles LLM recommendation circuits into forget-specific, retain-specific, and task-shared modules with tailored update rules to achieve more effective unlearning than weighted baselines.
Title resolution pending
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 3
method 1
citation-polarity summary
years
2026 6representative citing papers
PrecisionDiff is a differential testing framework that uncovers widespread precision-induced behavioral disagreements in aligned LLMs, including safety-critical jailbreak divergences across precision formats.
GenRecEdit injects cold-start items into generative recommendation models via context-aware token editing and interference-reducing triggers, boosting cold-start accuracy while using only 9.5% of retraining time.
TIGS detects backdoor-induced attention collapse in LLMs and applies content-aware tail-risk screening plus intrinsic geometric smoothing to suppress attacks while preserving normal performance.
Precise Shield identifies safety neurons in VLLMs via activation contrasts and aligns only them with gradient masking, boosting safety, preserving generalization, and enabling zero-shot cross-lingual and cross-modal transfer.
STEAR reduces spatial and temporal hallucinations in Video-LLMs via layer-aware evidence intervention from middle decoder layers in a single-encode pass.
citing papers explorer
-
Precise Shield: Explaining and Aligning VLLM Safety via Neuron-Level Guidance
Precise Shield identifies safety neurons in VLLMs via activation contrasts and aligns only them with gradient masking, boosting safety, preserving generalization, and enabling zero-shot cross-lingual and cross-modal transfer.
-
STEAR: Layer-Aware Spatiotemporal Evidence Intervention for Hallucination Mitigation in Video Large Language Models
STEAR reduces spatial and temporal hallucinations in Video-LLMs via layer-aware evidence intervention from middle decoder layers in a single-encode pass.