Towards Ethical Multi-Agent Systems of Large Language Models: A Mechanistic Interpretability Perspective
Preprint (arXiv) 2025
I am a postdoctoral researcher at the University of Hamburg, working on understanding and improving the reasoning mechanisms of modern neural networks to enhance their robustness and safety. As a principal investigator of the LUMO project, I focus on explaining, controlling, and composing the concepts learned by multimodal models to support reliable adaptation to new situations without catastrophic forgetting.
More information can be found on the Research, Teaching, and CV pages.
Research Interests
Latest News
- Oct 2025 Tutorial on Explaining Vision and Language Models: A Concept-Based Approach at ECAI 2025 (with Dr. Gesina Schwalbe).
- Oct 2025 Poster presentation about our recent paper at the Bonn Humboldt Award Winners’ Forum – The “what, how and why” of AI.
- Mar 2025 Workshop co-organizer for STRL 2025 (co-located with IJCAI 2025).
Recent Publications
-
-
Knowing the Facts but Choosing the Shortcut: Understanding How Large Language Models Compare Entities
Preprint (arXiv) 2025
-
Mental Modelling of Reinforcement Learning Agents by Language Models
Transactions on Machine Learning Research 2025
Recent Blog Posts
-
Deep Learning Courses and Books for Beginners
A curated list of deep learning resources for CS students who want to understand recent research papers.
-
Common LaTeX Mistakes by Newbies
A list of common LaTeX mistakes that beginners often make, with examples of correct usage.