一作ACL 2026Poster· Natural Language Processing
Among Us: Measuring and Mitigating Malicious Contributions in Model Collaboration Systems
Ziyuan Yang · Wenxuan Ding · Shangbin Feng · Yulia Tsvetkov
摘要
Language models (LMs) are increasingly used in collaboration: multiple LMs trained by different parties collaborate through routing systems, multi-agent debate, model merging, and more. Critical safety risks remain in this decentralized paradigm: what if some of the models in multi-LLM systems ar...