2 papers
cs.CR2026
Turn-Based Structural Triggers: Structure-Conditioned Backdoors in Multi-Turn LLMs
Yiyang Lu, Jinwen He, Yue Zhao +4
Large Language Models (LLMs) are increasingly deployed as multi-turn assistants and customized through instruction tuning with project-specific training components. This practice c…
cs.CR2026
CNT: Safety-oriented Function Reuse across LLMs via Cross-Model Neuron Transfer
Yue Zhao, Yujia Gong, Ruigang Liang +4
The widespread deployment of large language models (LLMs) calls for post-hoc methods that can flexibly adapt models to evolving safety requirements. Meanwhile, the rapidly expandin…