Skip to content
conftrace_

template region

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

inference-time attack (1) large language model (1) model vulnerability (1) mechanistic analysis (1) jailbreak attack (1) safety alignment (1) adversarial learning (1)

Papers

Why Safeguarded Ships Run Aground? Aligned Large Language Models’ Safety Mechanisms Tend to Be Anchored in The Template Region ACL 2025