You want to understand how things are connected and make a fundamental impact? We offer an environment where you can realize your full potential. At one of Europe’s largest and most modern business and economics universities. On a campus where quality of work is also quality of life. We are looking for support at the
Institute for Data, Process and Knowledge Management
Part-time, 30 hours/week
Starting September 01, 2026, and ending after 3.5 years
Qualified candidates with disabilities are especially encouraged to apply for this position. If necessary, a reduction of the extent of employment is possible.
Do you want to dive into a state-of-the-art research on large language modeling and information retrieval?
We are seeking a highly motivated prospective PhD student to join the newly established Vienna Research Group funded by WWTF and led by Dr. Svitlana Vakulenko. The objective of the group is to develop novel approaches to modeling knowledge specifically designed for use by an LLM. The group is positioned within the Institute for Data, Process and Knowledge Management, which boasts world-class excellence in reasoning, data management and knowledge representation research. We offer an environment where you can accelerate your learning outcomes, receive guidance and support to realize your full potential.
The ability of Large Language Models (LLMs) to generate contextually relevant natural language responses is truly impressive, and a growing number of people are using them on a regular basis to address their information needs. However, since LLMs are parametric models unlike databases, they were not designed to reliably store data. While they are trained on massive amounts of textual data from the Web and are likely to pick up factual knowledge from it, their responses are, while usually fluent and grammatically correct, often not factually correct, which makes them to appear plausible and effectively deceive their users. Alternative RAG approaches that condition LLM generation on a set of retrieved search results also fail in many cases, when information needs are more complex, and search results are not optimal.
