I’m a PhD student in the CopeNLU group at the University of Copenhagen, advised by Isabelle Augenstein . Before the PhD, I completed an MS at Copenhagen as well, joining the group in 2019.
My research sits at the intersection of Natural Language Processing and Computational Social Science. I’m interested in how large language models reflect and can shape our societies. Concretely, this has meant work on:
- Model bias and cultural values — how pre-trained and instruction-tuned models encode the values of the cultures whose text they were trained on, and what happens when they’re applied to populations they didn’t see.
- Online harms — stance detection, fact-checking, framing analysis, and the practical gap between what platforms actually need from this research and what the literature ends up producing.
- Equitable ML more broadly — fine-grained evaluations of where models fail, and methods for finding and removing biases once you can name them.
I’ve published at EMNLP, NAACL, ACL, EACL, AAAI, IJCAI, and in Computational Linguistics and ACM Computing Surveys. A full list of papers is on the papers page .
Outside of research, I try to spend time away from screens, mostly cooking new dishes or trying out a new restaurant.