Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
STAR: Detecting Inference-time Backdoors in LLM Reasoning via State-Transition Amplification Ratio
Seong-Gyu Park, Sohee Park, Jisu Lee +2
Recent LLMs increasingly integrate reasoning mechanisms like Chain-of-Thought (CoT). However, this explicit reasoning exposes a new attack surface for inference-time backdoors, whi…
cs.CL2025
Self-HarmLLM: Can Large Language Model Harm Itself?
Heehwan Kim, Sungjune Park, Daeseon Choi
Large Language Models (LLMs) are generally equipped with guardrails to block the generation of harmful responses. However, existing defenses always assume that an external attacker…