Lukas Ellinger

M. Sc.

Research Interests

With the rise of Large Language Models (LLMs), their tendency to produce false or misleading information has become a challenge for many users. My research focuses on improving the factuality and interpretability of LLMs to enhance their trustworthiness and societal impact. In my theses, I explore how NLP techniques can help verify and refine generated information, making it more reliable and accessible.

Current research topics include:

• Factual Correctness in NLP models
• Analyzing Stored Information in LLMs
• Reliable Response Generation
• Fact-Checking

Supervision of Theses

If you're interested in pursuing a thesis or guided research project related to my research interests, feel free to reach out via email! Please include your CV, Transcript of Records, and a brief (< 400 words) introduction about yourself and your motivation for the project.