About Me

I am currently pursuing my PhD at the Chair of Responsible Data Science under the supervision of Prof. Dr. Gjergji Kasneci. I hold a Master of Engineering degree from the University of Patras, Greece. For my Master’s thesis, I worked on the challenge of image captioning, implementing two distinct architectures. Beyond this, I have contributed in projects related to eXplainable AI frameworks, image inpainting applications, and recommender systems.

Research Focus

My research interests lie in the areas of AI Safety and Alignment. I am particularly focused on the problem of Unlearning, exploring methods to safely remove information from large language models. More broadly, I am interested in understanding how to control machine learning models with the ultimate goal of aligning them with human values and intentions.

Master Thesis Supervision Topics

Please contact me only if you are interested in the following topics:

1. Post-Training for AI Safety & Alignment
2. Machine Unlearning
3. Mechanistic Interpretability
4. Continual/Online Learning/Catastrophic Forgetting
5. Context Compression

Publications

A full list of my publications is available on my Google Scholar profile.