Formative Feedback in Times of Responsible Generative AI
In this webinar, I will explore the use of large language models (LLMs) in higher education, with a particular focus on generating formative feedback for student submissions. I will present the technical baselines for both student-facing and teacherfacing tools, discuss experiments examining their acceptance and the impact on student outputs. Furthermore, I will introduce our updated architecture for implementing an explainable, trustworthy, and responsible feedback system leveraging LLM capabilities. Finally, I will address the current limitations of this approach and outline opportunities for future research.
Breve CV: Dr. Leo Sylvio Rüdian received his Ph.D. in Computer Science from Humboldt University of Berlin, with a specialization in Learning Analytics and Educational Data Mining. He is currently a postdoctoral researcher at Humboldt University of Berlin, working on responsible generative AI in higher education, with a focus on LLM-based agents for formative feedback and trustworthy, compliant deployment. He serves as a reviewer for international conferences in the field of Artificial Intelligence in Education and is a multiple best paper award winner. His research interests include AI in educational technology, learning analytics, generative AI, adaptive and personalized learning, and responsible AI.
-
Leo S. Rüdian
