Goal / Research Questions:
- How can LLM outputs be decomposed into verifiable claims?
- How can claims be validated against trusted knowledge sources?
- How can hallucinations be detected and reduced?
Please contact Andreas Ekelhart.
The following contains currently open thesis topics in our areas of research. If you are interested in a topic or have a thesis idea of your own that might fit our research interests, please contact us directly.
Trustworthy and Verifiable LLM Reasoning for Cybersecurity
LLMs often produce outputs that are difficult to verify. This topic investigates methods for improving trustworthiness through structured validation and grounding in trusted knowledge sources.