conftrace_

prompt calibration

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

demonstration shortcut (1) mirror crafting (1) semantic safety (1) entropy guidance (1) adaptive defense (1) large language model (1) jailbreak defense (1) in-context learning (1) few-shot learning (1)

Papers

MirrorShield: Towards Dynamic Adaptive Defense Against Jailbreaks via Entropy-Guided Mirror Crafting AAAI 2026 Rectifying Demonstration Shortcut in In-Context Learning NAACL 2024