Do Pre-Trained Language Models Detect and Understand Semantic Underspecification? Ask the DUST!

arXiv:2402.12486v1 Announce Type: new
Abstract: In everyday language use, speakers frequently utter and interpret sentences that are semantically underspecified, namely, whose content is insufficient to fully convey their message or interpret them univocally. For example, to interpret the underspecified sentence "Don't spend too much", which leaves implicit what (not) to spend, additional linguistic context or outside knowledge is needed. In this work, we propose a novel Dataset of semantically Underspecified Sentences grouped by Type (DUST) and use it to study …

abstract arxiv cs.cl dust example language language models semantic speakers spend them type