NLSR patches only the safety-critical neurons in LoRA weights that changed after harmful fine-tuning, using a pre-amplified reference model, to restore safety without retraining.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
NLSR: Neuron-Level Safety Realignment of Large Language Models Against Harmful Fine-Tuning
NLSR patches only the safety-critical neurons in LoRA weights that changed after harmful fine-tuning, using a pre-amplified reference model, to restore safety without retraining.