Short Bio

Anastasios holds an M.Eng. degree in Electrical and Computer Engineering from Aristotle University of Thessaloniki. During his diploma thesis, he conducted research on efficient computer vision solutions for natural disaster management under the supervision of Professor I. Pitas. He has also worked as an R&D intern at Net2Grid, focusing on electricity theft detection using machine learning. Currently, he is a PhD fellow jointly affiliated with the Archimedes Research Unit and the University of Crete, where he works with Professor Nikos Komodakis on generative AI methods and their applications.

Research Interests

Anastasios’ research interests lie in generative artificial intelligence, with a particular focus on language modeling architectures, training objectives, and reasoning in Large Language Models.

PhD research on "Multi-Token Prediction Needs Registers"
Abstract: We propose MuToR, a simple and effective approach that leverages register tokens to predict future targets and enrich supervision for autoregressive transformers. Our method introduces only a negligible number of additional parameters and requires no architectural changes—ensuring compatibility with off-the-shelf pretrained language models.