Skip to content
conftrace_

layer-specific editing

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

safety layer (1) large language model (1) harmful response (1) prompt injection (1) adversarial prompt (1) jailbreak attack (1) model alignment (1) model safety (1)

Papers

Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing EMNLP 2024