Ruhr-Uni-Bochum

Elisabeth Kirsten receives 2026 Google PhD Fellowship

The program recognizes her outstanding research at the intersection of Machine Learning and Human-Centered AI

Young woman with curly hair and brown jacket smiling in front of a blurred background with trees

Elisabeth Kirsten is selected as one of the 2026 Google PhD Fellows in the category “Natural Language Processing”. She recently started her third year as a PhD student at the Chair of AI and Society at the Faculty of Computer Science and is affiliated with the UA Ruhr Research Center for Trustworthy Data Science and Security. Her research investigates the behaviour of Large Language Models (LLMs) in real-world situations.

The program supports exceptional graduate students who represent the future of computer science and related fields through their pioneering research. The fellowship provides vital financial support for their PhD pursuits and connects each Fellow with a global peer network.

In her research, Elisabeth Kirsten approaches AI from both technical and societal perspectives, aiming to understand how LLMs behave in real-world settings and how they affect individuals and society. As these systems increasingly move beyond answering questions to searching the web, interacting with tools, remembering information about users, and carrying out complex tasks, new questions arise about their reliability and societal impact. Elisabeth investigates these questions by developing methods to understand, evaluate and improve LLM-driven systems.

A central focus of her research is evaluating LLMs beyond traditional benchmark leaderboards, which measure performance on narrowly defined tasks and thus provide limited insight into how models behave in complex, open-ended and ambiguous real-world situations. Such situations often require AI systems to search for information, use external tools, and work through multiple steps to arrive at a solution. Elisabeth therefore aims to bridge the gap between benchmark performance and real-world behavior by investigating not only whether AI systems produce correct answers, but also how they arrive at them, when and why they fail, and which information they rely on.

She has already presented her research at prestigious international conferences, including ACL, COLM, NAACL, and WWW. Her research paper “Characterizing Web Search in The Age of Generative AI” received international media attention beyond academia.

“This fellowship recognizes Elisabeth’s research vision and her ability to connect human-centered considerations with rigorous ML research. Her pioneering work on generative search and her research on translating ideas from humanities and social sciences about knowledge diversity into LLM evaluation exemplify the breadth and technical depth of her approach”, highlights Professor Bilal Zafar, head of the AI and Society Chair and Elisabeth Kirsten’s supervisor. Her research is also supervised by Professor Nicole Krämer from University of Duisburg-Essen.


Since 2009, the Google PhD Fellowship Program has supported more than 1,100 graduate students worldwide. Over the past 17 years, alumni of the program have gone on to become tenured professors, startup founders, open-source pioneers, and research leaders helping shape the future of technology. In 2026, 148 new PhD fellows across 38 countries have been selected to receive the fellowship through $13 million in funding from Google.org.

 

General note: In case of using gender-assigning attributes we include all those who consider themselves in this gender regardless of their own biological sex.