Project IVAN

The research project IVAN is concerned with developing novel methods for detecting backdoors in software. As part of the project, we aim at bringing together concepts from computer security, program analysis, and machine learning to automatically identify unusual structures in code. The project will give rise to novel approaches for spotting backdoors, such as manipulated authentication, illegitimate control flows and elevated privileges. While the general detection of backdoors is an unsolvable (undecidable) problem, we are optimistic to raise the bar for adversaries to slip their code into legitimate software.

Further Information

The project officially started in April 2020 and ended in December 2023. It was funded by the German Federal Ministry of Education and Research. The project consortium consists of SAP Security Research and TU Braunschweig. Additional information are available in German on the corresponding webpage of the German Federal Ministry of Education and Research (BMBF).

Publications

• I still know it's you! On Challenges in Anonymizing Source Code.
  Micha Horlboge, Erwin Quiring, Roland Meyer and Konrad Rieck. Proceedings on Privacy Enhancing Technologies (PETS), 2024.
• PAVUDI: Patch-based Vulnerability Discovery using Machine Learning.
  Tom Ganz, Erik Imgrund, Martin Härterich and Konrad Rieck. Proc. of the 39th Annual Computer Security Applications Conference (ACSAC), 2023.
• Broken Promises: Measuring Confounding Effects in Learning-based Vulnerability Discovery.
  Erik Imgrund, Tom Ganz, Martin Härterich, Niklas Risse, Lukas Pirch and Konrad Rieck. Proc. of the 16th ACM Workshop on Artificial Intelligence and Security (AISEC), 2023.
• Hunting for Truth: Analyzing Explanation Methods in Learning-based Vulnerability Discovery.
  Tom Ganz, Philipp Rall, Martin Härterich and Konrad Rieck. Proc. of the 8th IEEE European Symposium on Security and Privacy (EuroS&P), 2023.
• CodeGraphSMOTE: Data Augmentation for Vulnerability Discovery.
  Tom Ganz, Erik Imgrund, Martin Härterich and Konrad Rieck. Proc. of the IFIP Conference on Data and Applications Security and Privacy (DBSEC), 2023.
• Detecting Backdoors in Collaboration Graphs of Software Repositories.
  Tom Ganz, Inaam Ashraf, Martin Härterich and Konrad Rieck. Proc. of the 14th ACM Conference on Data and Applications Security and Privacy (CODASPY), 2023.
• No Keys to the Kingdom Required: A Comprehensive Investigation of Missing Authentication Vulnerabilities in the Wild.
  Manuel Karl, Marius Musch, Guoli Ma, Martin Johns and Sebastian Lekies. Proc of the 22nd Internet Measurement Conference (IMC), 2022
• I still know it's you! On Challenges in Anonymizing Source Code.
  Micha Horlboge, Erwin Quiring, Roland Meyer and Konrad Rieck. Technical Report (arXiv:2208.12553), 2022.
• Explaining Graph Neural Networks for Vulnerability Discovery.
  Tom Ganz, Martin Härterich, Alexander Warnecke and Konrad Rieck. Proc. of the 14th ACM Workshop on Artificial Intelligence and Security (AISEC), November 2021. (Best Paper Award)
• TagVet: Vetting Malware Tags using Explainable Machine Learning.
  Lukas Pirch, Alexander Warnecke, Christian Wressnegger and Konrad Rieck. Proc. of the 14th ACM European Workshop on Systems Security (EuroSec), 2021.