(Large) Language Models achieve impressive results on various NLP tasks. LLMs are, however, based on purely distributional data; they still rely on learning (possibly complex and sophisticated) associations between words and their contexts. It is difficult to tell to what degree language models can reflect a human-like understanding of semantics and how specific information is reflected. Can models perform interpretation and abstraction to such a degree that they can solve analogies? Given that models are trained on large amounts of text, what social perspectives do they reflect? Do they engage in stereotyping? And finally, how do we talk about and think about language models and their abilities? In this talk, I will present findings from several studies investigating the behavior of (large) language models and discuss them with respect to an analysis of the language we use to talk about AI.

Pia Sommerauer

Pia Sommerauer is an assistant professor at VU Amsterdam. Her research focuses on understanding how people use language to form (social) categories and on how language models represent social and semantic categories. She approaches these questions from interdisciplinary perspectives and collaborates with experts in fields such as communication science and philosophy. She teaches courses in programming and Natural Language Processing to diverse and interdisciplinary groups of students.

Social Media

Bluesky

LinkedIn

Personal Website