research

research interests.

Making a positive impact is what drives me. Hence, the following research topics and questions are (amongst others) what my current interests are and what I want to work on in the future:

Understanding generalization capabilities. This includes rigorous analyses of model capabilities and extensive model evaluation; designing tools and metrics that go beyond aggregate metrics to give us more insight into model strengths and weaknesses. What is in the data? What did the model learn from it? What does this mean for real-world data?

AI for social good. Language models are increasingly affecting our daily lives and especially certain applications require us to develop models with a lot of care. This goes beyond the conventional performance-driven model development. How can we address such safety-critical applications?

Multicultural Human-Centred systems design. As end-users of most language models, it is of utmost importance that the systems are designed with keeping humans in mind. Every user comes with their own experience, values, and beliefs. How can we design adaptive systems that accommodate this diversity and subjectivity? How can we ensure that our systems are useful for a wide set of users?