1 paper
Jiajie He, Jiangyuan Hong, Dongling Ni +2
Large language models (LLMs) are increasingly used as user simulators, yet it remains unclear whether their predictions faithfully reproduce the evolving decisions of individual us…