I have authored several reseach papers on privacy-preserving and explainable ML.
Please also check out my GitHub for the corresponding implementation packages and Google Scholar.
Tobias Leemann, Periklis Petridis, Giuseppe Vietri, Dionysis Manousakas, Aaron Roth, Sergul Aydore. International Conference on Learning Representations (ICLR), 2025.
Attention Mechanisms Don’t Learn Additive Models: Rethinking Feature Importance for Transformers
Tobias Leemann, Alina Fastowski, Felix Pfeiffer, and Gjergji Kasneci.Transactions on Machine Learning Research (TMLR), 2024.
I Prefer not to Say: Operationalizing Fair and User-guided Data Minimization
Tobias Leemann, Martin Pawelczyk, Christian Thomas Eberle, and Gjergji Kasneci. AAAI Conference on Artificial Intelligence, 2024.
Gaussian Membership Inference Privacy
Tobias Leemann*, Martin Pawelczyk*, and Gjergji Kasneci. Advances in Neural Information Processing Systems (NeurIPS), 2023.
Language Models are Realistic Tabular Data Generators
Vadim Borisov, Kathrin Seßler, Tobias Leemann, Martin Pawelczyk, and Gjergji Kasneci.International Conference on Learning Representations (ICLR), 2023.
When Are Post-hoc Conceptual Explanations Identifiable?
Tobias Leemann,* Michael Kirchhof,* Yao Rong, and Gjergji Kasneci. Uncertainty in Artificial Intelligence (UAI), 2023.
Deep Neural Networks and Tabular Data: A Survey
Vadim Borisov, Tobias Leemann, Kathrin Seßler, Johannes Haug, Martin Pawelczyk, and Gjergji Kasneci. IEEE Transactions on Neural Networks and Learning Systems, Early Access, 2022.
A Consistent and Efficient Evaluation Strategy for Attribution Methods
Yao Rong,* Tobias Leemann,* Vadim Borisov, Gjergji Kasneci, and Enkelejda Kasneci. International Conference on Machine Learning, 2022.
Multi-Step Training for Predicting Roundabout Traffic Situations
Moritz Sackmann, Tobias Leemann, Henrik Bey, Ulrich Hofmann, and Jörn Thielecke. IEEE International Intelligent Transportation Systems Conference (ITSC), 2021.
Distribution Preserving Multiple Hypotheses Prediction for Uncertainty Modeling
Tobias Leemann, Moritz Sackmann, Jörn Thielecke, and Ulrich Hofmann. 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), 2021.
Tobias Leemann. Dissertation. Universität Tübingen, 2025.