Investigating grammatical abstraction in language models using few-shot learning of novel noun gender

arXiv:2403.10338v1 Announce Type: new
Abstract: Humans can learn a new word and infer its grammatical properties from very few examples. They have an abstract notion of linguistic properties like grammatical gender and agreement rules that can be applied to novel syntactic contexts and words. Drawing inspiration from psycholinguistics, we conduct a noun learning experiment to assess whether an LSTM and a decoder-only transformer can achieve human-like abstraction of grammatical gender in French. Language models were tasked with learning the gender …

abstract abstraction agreement arxiv cs.cl examples few-shot few-shot learning gender humans inspiration language language models learn notion novel rules type word words