BIFOLD supports the 4th IEEE Conference on Secure and Trustworthy Machine Learning (SaTML 2026), held March 23 - 25, 2026, at the Technical University of Munich, Germany. The conference brings together researchers to advance the theoretical and practical understanding of vulnerabilities in machine learning and to foster the development of robust, trustworthy machine learning systems. Topics range from novel attacks and defenses to privacy, fairness, and interpretability.
Contributions
BIFOLD researchers contribute to this year's program: Konrad Rieck, chair of the Machine Learning and Security group, serves as Program Chair alongside Rachel Cummings of Columbia University. The Web Chairs, Thorsten Eisenhofer and Stefan Czybik, contribute to the conference's technical infrastructure. Thorsten Eisenhofer and Kiril Bykov are additionally part of the Program Committee, which spans a large international community of researchers.
On Wednesday, March 25, the paper “Beyond the TESSERACT: Trustworthy Dataset Curation for Sound Evaluations of Android Malware Classifiers” will be presented. The paper is co-authored by BIFOLD researcher Lorenz Linhardt. The work, in collaboration with researchers from King's College London, University College London, HiddenLayer, and TU Wien, identifies five novel factors that affect the reliability of Android malware classifier evaluations and proposes a methodology for trustworthy dataset curation.
Paper: https://discovery.ucl.ac.uk/id/eprint/10220473/1/chow-satml26.pdf
Authors: Theo Chow, Mario D'Onghia, Lorenz Linhardt, Zeliang Kan, Daniel Arp, Lorenzo Cavallaro, Fabio Pierazzi