Topinformationen

My research

Many Artificial Intelligence models have long been considered as “black box”- referring to the fact that it is unclear how such models come to their results, no matter how significant they are. Explainable AI has long been sought, especially in high-stake situations like medical applications, autonomous cars, or bank loans. Here, the reasons behind a decision come to be as important as the decision itself.

With the rapid rise in Large Language Models (LLMs), and the direct effect it has on our lives, it’s equally important to investigate their internal mechanism and potential biases. That is why I’m currently focusing on explainability of LLMs, and whether there is a trade-off between the explainability of a model and its performance/creativity.

Teaching

Winter term 2024/25

• Explainability of Large Language Models
• Methods of AI (Seminar)

Summer term 2024

• Computational Creativity in Natural Language Processing