We are a group of researchers investigating machine learning, language, and the human mind. We are based at Saarland University and Saarland Informatics Campus, Germany.
Our research studies the following topics:
Abilities of Language Models, and how they are learned:We investigate the abilities and learning biases of transformer models and other architectures, using both theoretical and empirical methods (ICLR 2025, ICML 2025, ACL 2024, NeurIPS 2024, TACL 2020).
Understanding Inner Workings of Language Models:We develop methods for mechanistically reverse-engineering the inner workings of language models (COLM 2025, NeurIPS 2024, TACL 2019).
Computational Cognition and Neuroscience: We use machine learning to model human language proecessing (PNAS 2022a, PNAS 2020, PNAS 2022b) and human perception (Nature Neuroscience 2024).