Leon Eshuijs

I am a PhD student at the Computational Linguistics and Text Mining Lab (CLTL) at the Vrije Universiteit Amsterdam. My research is focused on Reinforcement Learning for Natural Language Processing and is part of the Hybrid Intelligence Center.

My research interests lie in aligning AI models with human values to ensure their safety. For the same reason, I am interested in using Mechanistic Interpretability to improve our understanding of how Large Language Models work, thereby ensuring their safety and detecting failed cases of undesirable behavior. My supervisors are Antske Fokkens and Shihan Wang.

Before starting my PhD, I obtained my Master’s degree in Artificial Intelligence from the University of Amsterdam, where I focused on Deep Learning. For my Master’s thesis, I investigated Knowledge Injection on Transformers for image captioning. Afterward, I completed a semester in Physics at Sorbonne Université as part of my Erasmus, building on the Physics courses I took as part of my bachelor’s honors program. I am also passionate about applying AI to real-world problems. In the past, I have worked on a robotic quadruped project (see projects).