Andrej Kastrin

Associate Professor at University of Ljubljana

• Email |
• Google Scholar |
• LinkedIn |
• GitHub

Research

My current research focuses on natural language processing, machine learning, large language model, and explainable AI. Our goal is to uncover the mechanisms behind large language model and use that understanding to build trustworthy models that are reliable, truthful, and safe. Specifically, we are interested in the following research topics:

1. Mechanistic Study of LLMs: How can we open the black box of LLMs to uncover their internal mechanisms, especially those enabling complex reasoning?
2. Trustworthy LLMs Guided by Mechanism: How can insights from mechanistic understanding be translated into practice for building LLMs that are reliable, truthful, and safe in real-world applications?

Recent News