Prof. Dr. Stefan Bauer
Academic Career and Research Areas
Our team develops approaches that enable models to refine their internal hypotheses, adapt their computation to the task at hand, and work with discrete, structured representations. These capabilities strengthen a model’s ability to form abstractions, perform adaptive inference, and carry out multi-step decision making. Our goal is to advance AI systems that can address increasingly complex reasoning tasks with greater flexibility and precision, and in doing so, help uncover the underlying principles of intelligence.
Before joining TUM, Stefan Bauer was an assistant professor at KTH Stockholm, a group leader at the Max Planck Institute for Intelligent Systems in Tübingen and a Visiting Researcher at MILA, GSK and Microsoft Research. He obtained his Ph.D. in Computer Science from ETH Zurich. Since 2025, he has been a CIFAR LMB Fellow and co-coordinates the Helmholtz Foundation Model Initiative, one of Germany’s major national efforts for developing large-scale AI models.
Awards
- CIFAR Fellowship in Learning in Machines and Brains Program (2025)
- Best paper International Conference for Machine Learning (ICML) (2019)
- ETH medal for outstanding doctoral thesis (2018)
Key Publications (all publications)
V. Pauline, T. Höppe, K. Neklyudov, A. Tong, S. Bauer, A. Dittadi 2025. Foundations of Diffusion Models in General State Spaces: A Self-Contained Introduction.
AbstractO. Ahmed*, F. Träuble*, A. Goyal, A. Neitz, M. Wuthrich, Y. Bengio, B. Schölkopf, S. Bauer, 2020. CausalWorld: A Robotic Manipulation Benchmark for Causal Structure and Transfer Learning, International Conference on Learning Representations (ICLR).
AbstractF. Kapl*, E. Angelis*, T. Höppe*, K. Maile٭, J. von Oswald٭, N. Scherrer٭, S. Bauer٭. Do Depth-Grown Models Overcome The Curse Of Depth? An In-Depth Analysis arxiv 2025.
AbstractP. Tigas, Y. Annadani, A. Jesson, B. Schölkopf, Y. Gal and S. Bauer. "Interventions, Where and How? Experimental Design for Causal Models at Scale“. Neural Information Processing Systems (NeurIPS) (2022).
AbstractIf you wish your profile to be changed or updated please contact Franz Langer.