Skip to content
conftrace_

llm vulnerability

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

poison pill attack (1) emoticon semantic confusion (1) prompt-based mitigation (1) silent failure (1) knowledge retrieval (1) inference-time alignment (1) best-of-n sampling (1) jailbreak attack (1) adversarial attack (1) model pruning (1)

Papers

False Friends in the Shell: Unveiling the Emoticon Semantic Confusion in Large Language Models ACL 2026 Jailbreaks as Inference-Time Alignment: A Framework for Understanding Safety Failures in LLMs EACL 2026 Swallowing the Poison Pills: Insights from Vulnerability Disparity Among LLMs IJCNLP 2025