Tag
This paper applies philosophy of science to argue that LLMs offer epistemic value as minimal models for how-possibly explanations in linguistics, but do not yet qualify as how-actually explanations of human language.