Seminar: "Modelling the Language of Extremist Groups"
Abstract
Extremist groups develop complex in-group language, also referred to as cryptolects, to exclude or mislead outsiders. Though this is a longstanding and well-documented social norm, it highlights key shortcomings in current natural language processing technologies, e.g. large language models (LLMs), especially when used for content moderation. In this talk, I will describe recent methods, datasets, and models we developed to address these challenges. Our experiments center on two online extremist platforms, Incels and Stormfront, which promote alt-right and misogynistic ideologies, respectively.
Invited Speaker: Christine de Kock (University of Melbourne)
Bio: Dr Christine de Kock is an assistant professor in the School for Computing and Information Systems at the University of Melbourne. Her research explores topics in computational sociolinguistics, with a particular focus on harmful online behaviour. She completed her PhD in the Natural Language and Information Processing group at Cambridge.
