Sebastian Szyller

Sebastian Szyller

Assistant Professor
Department of Computer Science

My name is Sebastian Szyller and I'm a tenure track assistant professor at the Department of Computer Science at Aalto University. I lead new Trustworthy and Adversarial Computing Lab. I'm interested in different ways that we can protect machine learning models and data to enable secure and trustworthy analysis -- both in terms of the technical details as well as legislation compliance.

Prior to returning to Aalto University I was a research scientist at Intel Labs, where I worked on provenance & ownership, privacy and adversarial robustness in generative, multimodal machine learning systems.

I did my MSc and PhD as part of the doctoral track which is a joint MSc and PhD programme at Aalto University. Both my graduate and doctoral studies (including thetheses) were supervised by N. Asokan. My doctoral dissertation received Finnish AI Society Best Dissertation Award, and Aalto University Best Dissertation Award.

I did my BSc in computer science at Lodz University of Technology (and briefly Turku University of Applied Sciences ), and wrote my thesis under the supervision of Laurent Babout. I was also a member (and eventually a vice-president) of the student government at my faculty, a member of the scholarship committee, and I organised language workshops for students. In parallel, I worked as a big data and Scala software engineer at an investment bank.

In my free time, I enjoy film & street photography, custom mechanical keyboards, industrial design, bouldering and fixed-gear cycling. In the previous life, I was a member of a British Parliamentary style debate club Aalto Debating Society.

Full researcher profile
https://research.aalto.fi/...
Postal address
Konemiehentie 2 02150 Espoo Finland
Phone number
+358504633054

Publications

Soft Token Attacks Cannot Reliably Audit Unlearning in Large Language Models

Haokun Chen, Sebastian Szyller, Weilin Xu, Nageen Himayat 2025 Findings of the Association for Computational Linguistics: EMNLP 2025

Atlas: A Framework for ML Lifecycle Provenance & Transparency

Marcin Spoczynski, Marcela S. Melara, Sebastian Szyller 2025 Proceedings - 10th IEEE European Symposium on Security and Privacy Workshops, Euro S and PW 2025

Amulet: a Python Library for Assessing Interactions Among ML Defenses and Risks

Asim Waheed, Vasisht Duddu, Rui Zhang, Sebastian Szyller, N. Asokan 2025

SoK : Unintended Interactions among Machine Learning Defenses and Risks

Vasisht Duddu, Sebastian Szyller, N. Asokan 2024 Proceedings - 45th IEEE Symposium on Security and Privacy, SP 2024

False Claims against Model Ownership Resolution

Jian Liu, Rui Zhang, Sebastian Szyller, Kui Ren, N. Asokan 2024 Proceedings of the 33rd USENIX Security Symposium

LLM Self Defense: By Self Examination, LLMs Know They Are Being Tricked

Mansi Phute, Alec Helbling, Matthew Daniel Hull, ShengYun Peng, Sebastian Szyller, Cory Cornelius, Duen Horng Chau 2024 The Second Tiny Papers Track at ICLR 2024

Imperceptible Adversarial Examples in the Physical World

Weilin Xu, Sebastian Szyller, Cory Cornelius, Luis Murillo Rojas, Marius Arvinte, Alvaro Velasquez, Jason Martin, Nageen Himayat 2024 arXiv.org

Conflicting Interactions among Protection Mechanisms for Machine Learning Models

Sebastian Szyller, N. Asokan 2023 AAAI-23 Special Tracks

On the Robustness of Dataset Inference

Sebastian Szyller, Rui Zhang, Jian Liu, N Asokan 2023 Transactions on Machine Learning Research