SafeAnchor preserves 93.2% of original safety alignment across sequential domain adaptations by anchoring low-rank safety subspaces and constraining orthogonal updates, while matching unconstrained fine-tuning performance within 1.5 points.
Safe lora: the silver lining of reducing safety risks when fine-tuning large language models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.
citing papers explorer
-
SafeAnchor: Preventing Cumulative Safety Erosion in Continual Domain Adaptation of Large Language Models
SafeAnchor preserves 93.2% of original safety alignment across sequential domain adaptations by anchoring low-rank safety subspaces and constraining orthogonal updates, while matching unconstrained fine-tuning performance within 1.5 points.
-
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.