1 paper
Oscar Lithgow-Serrano, Vani Kanjirangat, Alessandro Antonucci
Recent papers show LLMs achieve near-random accuracy in causal relation classification, raising questions about whether such failures arise from limited pretraining exposure or dee…