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

1 code implementation • 19 Feb 2024 • Frank Wildenburg, Michael Hanna, Sandro Pezzelle

In this work, we propose a novel Dataset of semantically Underspecified Sentences grouped by Type (DUST) and use it to study whether pre-trained language models (LMs) correctly identify and interpret underspecified sentences.

Sentence