Luca Modica

I’m a Deep Learning Researcher at Zenseact, working on Multimodal Large Language (Action) models for end-to-end planning in autonomous driving. Previously, I completed my master’s degree in Data Science and AI at Chalmers University of Technology, conducting a master’s thesis on mechanistic interpretability and factuality of multimodal speech-language models with the NLP@DSAI group. I have also been part of the social computing group at University of Turin, where I collaborated as research fellow for the FirstLife civic social network project.

My current research interests include Vision Language-Action Models, (mechanistic) interpretability of deep learning models, and their applications in autonomous driving. In my free time, you will probably find me running, bouldering, traveling✈️, or exploring the beutiful nature of Sweden🌲.