SOURCE-LINKED INTELLIGENCE

NeuronGuard: Robust LLM Safety Alignment via Ablation-Aware Safety Signal Redistribution

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Safety alignment in large language models (LLMs) remains brittle against a growing spectrum of attacks. Jailbreak attacks bypass safety mechanisms through crafted prompts, while neuron-level attacks directly prune safety-critical neurons post-deployment. Both exploit a common weakness: safety-relevant information concentrates in a sparse neuron subset. We present NeuronGuard, a fine-tuning-stage defense that simultaneously hardens LLMs against both attack classes by redistributing safety signals across a broader set of neurons. NeuronGuard dynamically identifies safety-critical neurons via per

Read original source ↗ Open in workspace

recordType

paper

region

Global