Moa Johansson is interested in several aspects of AI, including using large language models for application not only in NLP but also for program and specification synthesis, and for formalized mathematics. She is also interested in NLP for social science applications, and methods for combining neural and symbolic AI (neuro-symbolic AI).

I am interested in natural language processing: in our research group, we develop machine learning models and algorithms that analyse language written by humans. I have worked on variety of research problems in NLP and my interests are quite wide. Most of the current research in our group involves questions around learned representations of words and text: how we can measure what information is expressed by language representations, and how we can control it.

Joined the lab on 2016-04-01

I work in the field of natural language processing and political science. My interests are in representation and explainability within language models and how that may contribute to the study of political behaviour.

Joined the lab on 2020-09-01

I’m interested in language model controllability, interpretability and prediction provenance. In my research I investigate factual knowledge in language models and methods for adding knowledge to a model.

Joined the lab on 2020-09-01

I am interested in using natural language processing in political and social science research.

Joined the lab on 2022-01-01

I focus my research on controlling and interpreting language models within Conversational AI.

Joined the lab on 2022-01-10

I’m interested in causal inference and natural language processing. In my research I develop methods to use text data for inferring causal estimates, and apply those methods in research questions from the social sciences. I am supervised by Richard Johansson and co-supervised by Moa Johansson and Adel Daoud.

Joined the lab on 2023-09-06