Skip to content
conftrace_

safety neuron

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

self-attention layer (1) model tuning (1) jailbreak attack (1) adversarial defense (1) model alignment (1) safety alignment (1) instruction tuning (1)

Papers

Unraveling LLM Jailbreaks Through Safety Knowledge Neurons EACL 2026 Understanding and Enhancing Safety Mechanisms of LLMs via Safety-Specific Neuron ICLR 2025