Philipp Normann
MSc
Research Areas
- Cybersecurity, Machine Learning
Role
-
PreDoc Researcher
Security and Privacy, E192-06
Publications
-
Towards Explaining Classification Models in Security with Sparse Autoencoders
/
Breuer, N. O., Linhardt, L., Normann, P., & Arp, D. (2026). Towards Explaining Classification Models in Security with Sparse Autoencoders. In 2026 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW) (pp. 278–289). IEEE. https://doi.org/10.1109/EuroSPW72509.2026.00041
Project: BREADS (2024–2030) -
Chasing Shadows: Pitfalls in LLM Security Research
/
Jonathan Evertz, Risse, N., Neuer, N., Müller, A., Normann, P., Sapia, G., Gupta, S., Pape, D., Shaw, S., Srivastav, D., Wressnegger, C., Quiring, E., Eisenhofer, T., Arp, D., & Schönherr, L. (2026). Chasing Shadows: Pitfalls in LLM Security Research. In Proceedings 2026 Network and Distributed System Security Symposium. Network and Distributed System Security (NDSS) Symposium 2026, San Diego, CA, United States of America (the). Schloss Dagstuhl. https://doi.org/10.14722/ndss.2026.241749
Project: BREADS (2024–2030) - Identifying Offline Metrics that Predict Online Impact: A Pragmatic Strategy for Real-World Recommender Systems / Wilm, T., & Normann, P. (2025). Identifying Offline Metrics that Predict Online Impact: A Pragmatic Strategy for Real-World Recommender Systems. In RecSys ’25: Proceedings of the Nineteenth ACM Conference on Recommender Systems (pp. 967–970). Association for Computing Machinery. https://doi.org/10.1145/3705328.3748111
Supervisions
-
User behavior simulation with large language models for security evaluation
/
Moser, L. (2026). User behavior simulation with large language models for security evaluation [Diploma Thesis, Technische Universität Wien]. reposiTUm. https://doi.org/10.34726/hss.2026.132647
Download: PDF (6.22 MB)