Vera Rimmer

Research Expert at DistriNet, KU Leuven, Belgium

prof_pic.jpg

I conduct and lead research activities at the intersection of security, privacy, and AI. Our team explores data analytics in network intrusion and malware detection, privacy-enhancing technologies, and trustworthiness of data-driven AI in the wider ICT context. The driving force behind my work is examining the optimal role of technology, particularly AI, in our society while balancing its benefits and challenges. In this age of surveillance and uncontrolled data collection and inference, I am committed to developing a comprehensive understanding, forming reasonable expectations, and devising effective strategies to mitigate the risks of applied AI.

Beyond academic research, our team is committed to sharing scientific knowledge through direct collaboration with industry. We work with companies seeking practical guidance on how to safely and securely leverage modern AI technologies, such as deep learning, foundation models, and large language models, to meet their needs.

What's new?

Jun 04, 2025 Image I returned to SecAppDev 2025 with two talks for practitioners: on “Navigating the Security Landscape of Modern AI”, and on “The Engineer’s Guide to Data Privacy”. The first talk was also repeated for Cyber Security Coalition: Application Security Experience Sharing Day.
May 19, 2025 Image Our team published a pre-print of our research on adversarial purification: our method called FlowPure based on continuous normalizing flows outperforms state-of-the-art purifiers. We will present a poster at IEEE EuroS&P 2025 in Venice!
May 14, 2025 Image I had a pleasure to give a guest lecture on “Privacy Engineering Technologies” for the Data Application and Security course at the University of Liechtenstein.
Jan 22, 2025 Image I participated in the NDC Security 2025 conference in Oslo, where I gave a talk on Nagivating the Security and Privacy Landscape of Modern AI.
Dec 20, 2024 Image Our 4th Workshop on Rethinking Malware Analysis (WoRMA) is accepted to appear at IEEE EuroS&P 2025 in Vienna, Austria! Co-chaired with Fabio Pierazzi and Simone Aonzo.

Selected publications

  1. arXiv
    FlowPure: Continuous Normalizing Flows for Adversarial Purification
    Elias Collaert, Abel Rodrı́guez, Sander Joos, Lieven Desmet, and Vera Rimmer
    arXiv preprint arXiv:2505.13280, 2025
  2. SatML
    SoK: On the offensive potential of AI
    Saskia Laura Schröer, Giovanni Apruzzese, Soheil Human, Pavel Laskov, Hyrum S Anderson, Edward WN Bernroider, and 10 more authors
    In 3rd IEEE Conference on Secure and Trustworthy Machine Learning, 2025
  3. arXiv
    Adversarial Markov Games: On Adaptive Decision-Based Attacks and Defenses
    Ilias Tsingenopoulos, Vera Rimmer, Davy Preuveneers, Fabio Pierazzi, Lorenzo Cavallaro, and Wouter Joosen
    arXiv preprint arXiv:2312.13435, 2023
  4. PETS
    Trace Oddity: Methodologies for Data-Driven Traffic Analysis on Tor
    Vera Rimmer, Theodor Schnitzler, Tom Van Goethem, Abel Rodrı́guez Romero, Wouter Joosen, and Katharina Kohls
    Proceedings on Privacy Enhancing Technologies, 2022
  5. WoRMA
    Position Paper: On Advancing Adversarial Malware Generation Using Dynamic Features
    Ali Shafiei, Vera Rimmer, Ilias Tsingenopoulos, Lieven Desmet, and Wouter Joosen
    In Proceedings of the 1st Workshop on Robust Malware Analysis, 2022
  6. Springer
    Open-World Network Intrusion Detection
    Vera Rimmer, Azqa Nadeem, Sicco Verwer, Davy Preuveneers, and Wouter Joosen
    In Security and Artificial Intelligence: A Crossdisciplinary Approach, 2022
  7. WTMC
    Troubleshooting an intrusion detection dataset: the CICIDS2017 case study
    Gints Engelen, Vera Rimmer, and Wouter Joosen
    In 2021 IEEE Security and Privacy Workshops (SPW), 2021
  8. NDSS
    Automated Website Fingerprinting through Deep Learning
    Vera Rimmer, Davy Preuveneers, Marc Juarez, Tom Van
Goethem, and Wouter Joosen
    In Proceedings of the 25nd Network and Distributed System
Security Symposium (NDSS 2018), 2018
  9. WOOT
    Fishy faces: Crafting adversarial images to poison face authentication
    In 12th USENIX Workshop on Offensive Technologies (WOOT 18), 2018