I am a postdoctoral researcher in the Computational Linguistics Department at Heinrich Heine University (HHU), working on the DFG-funded project “Coercion and Copredication as Flexible Frame Composition”.

My research focuses on language modeling and representation learning, and their interaction with lexical semantics. I develop methods for better understanding contextualized word embeddings, using diverse approaches such as adversarial learning, contrastive learning, and interpretability and explainability methods. I also work on geometric analysis of the embedding space.