1 paper
Matthieu Tehenan, Christian Bolivar Moya, Tenghai Long +1
Large language models (LLMs) have demonstrated emergent abilities across diverse tasks, raising the question of whether they acquire internal world models. In this work, we investi…