Domain-Adapted Pre-trained Language Models for Implicit Information Extraction in Crash Narratives

Open in new window