Nils Loose 🏔
Nils Loose

Research Assistant/PhD Student

University of LĂŒbeck

About Me

I am a research assistant/PhD student at the Institute for IT-Security at the University of LĂŒbeck. I am intrigued by the intersection of machine learning and code analysis. I maintain the dynamic symbolic execution engine SWAT

Interests
  • Machine Learning
  • Adversarial Attacks
  • Code Analysis
Education

  • PhD Student

    University of LĂŒbeck

  • MSc in Computer Science

    University of LĂŒbeck

  • BSc in Computer Science

    University of LĂŒbeck

📚 My Research

I am a researcher at the Institute for IT Security at the University of LĂŒbeck, specializing in machine learning program analysis and cybersecurity. My work spans machine learning, adversarial attacks, and symbolic methods for understanding and securing software and models.

Please reach out to collaborate 😃

Featured Publications
SWAT: Modular dynamic symbolic execution for java applications using dynamic instrumentation (competition contribution)
Recent Publications
Felix MĂ€chtle, Nils Loose, Jan-Niclas Serr, Jonas Sander, Thomas Eisenbarth (2025). AutoStub: Genetic Programming-Based Stub Creation for Symbolic Execution. To be published at SBFT 2025.
Cite
Felix MĂ€chtle, Jan-Niclas Serr, Nils Loose, Jonas Sander, Thomas Eisenbarth (2024). OCEAN: Open-World Contrastive Authorship Identification. arXiv preprint arXiv:2412.05049.
Cite
Georg Becker, Thomas Eisenbarth, Hannes Federrath, Mathias Fischer, Nils Loose, Simon Ott, Joana Pecholt, Stephan Marwedel, Dominik Meyer, Jan Stijohann, Others (2024). SOVEREIGN-towards a holistic approach to critical infrastructure protection. Proceedings of the 19th International Conference on Availability, Reliability and Security.
Cite
Nils Loose, Felix MĂ€chtle, Florian Sieck, Thomas Eisenbarth (2024). SWAT: Modular dynamic symbolic execution for java applications using dynamic instrumentation (competition contribution). International Conference on Tools and Algorithms for the Construction and Analysis of Systems.
PDF Cite Code DOI
Nils Loose, Felix MĂ€chtle, Claudius Pott, Volodymyr Bezsmertnyi, Thomas Eisenbarth (2023). Madvex: Instrumentation-Based Adversarial Attacks on Machine Learning Malware Detection. International Conference on Detection of Intrusions and Malware, and Vulnerability Assessment.
Cite
Recent News