Abstract: The ability of Large Language Models (LLMs) to achieve impressive performances in natural language processing tasks derives from the large amounts of knowledge they recover from distributional statistics in huge textual corpora. However, the precise type, breadth and reliability of this knowledge is still an open issue. In this talk, I will present ongoing research that aims at testing the knowledge of LLMs about events, their typicality, and semantic relations.
Location: Attend in person at room J411 or via Zoom, https://gu-se.zoom.us/j/66299274809?pwd=Yjc2ejc2VVhraXVJMmhWeWtOQ2NuUT09
Time: 13:15-15:00