When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models

arXiv:2605.11612v1 Announce Type: cross Abstract: Backdoor vulnerabilities widely exist in the fine-tuning of large language models(LLMs). Most backdoor poisoning methods operate mainly at the token level and lack deeper semantic manipulation, which limits stealthiness. In addition, Prior attacks rely on a single fixed trigger to induce harmful outputs. Such static triggers are easy to detect, and clean fine-tuning can weaken the trigger-target association. Through causal validation, we observe