作者位次 #3/6ACL 2026Findings· Findings of ACL
User-Assistant Bias in LLM s
Xu Pan · Jingxuan Fan · Zidi Xiong · Ely Hahami · Jorin Overwiening · Ziqian Xie
摘要
Modern large language models (LLMs) are typically trained and deployed using structured role tags (e.g. system, user, assistant, tool) that explicitly mark the source of each piece of context. While these tags are essential for instruction following and controllability, asymmetries in the trainin...