作者位次 #9/14ACL 2026Findings· Findings of ACL
Multi-Hop Knowledge Editing via Critic-Guided Multi-Agent Reasoning
Xudong Li · Yuhang Tian · Dandan Song · Zhijing Wu · Shuhao Zhang · Jun Yang · Yongyu Huo · Changzhi Zhou · Xinyu Zhang · Chenhao Li · Huipeng Ma · Luan Zhang · Yan Xu · Qian Liu
摘要
Knowledge within large language models (LLMs) inevitably lags behind an evolving world, motivating knowledge editing methods that update facts without expensive retraining. In multi-hop knowledge editing, models must not only recall updated facts but also correctly propagate them through multi-st...