1 paper
Yujin Zhou, Mingxuan Zheng, Chuxue Cao +4
As large language models are increasingly deployed as tool-augmented legal agents, they introduce agentic hallucinations where tool-call and reasoning errors cascade into fabricate…