PhD position: Improving Language Models Formal Industry

Organisation/Company: KU LEUVEN Research Field Computer science » Informatics Researcher Profile First Stage Researcher (R1) Final date to receive applications 24 Aug 2026 - 23:59 (UTC) Country Belgium Type of Contract Temporary Job Status Full-time Offer Starting Date 1 Oct 2026 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Reference Number BAP
- Is the Job related to staff position within a Research Infrastructure? No

In industry, language models are currently being widely experimented with to build, among other things, intelligent assistants and company‑specific search functions. However, we notice that companies still have many questions about reliability: modern LLMs still do not guarantee correctness and tend to hallucinate facts. By contrast, other AI technologies, such as knowledge graphs and formal reasoning engines, are able to reason reliably, but are less good at handling ambiguity.

This PhD focuses on combining these two types of systems, in close collaboration with industry. Research into this kind of neuro‑symbolic AI is growing, but before it can be applied in industry, many obstacles still need to be overcome. Together with the companies in the consortium, you will examine their needs and questions, investigate what problems remain, and develop new methods and techniques to remove these barriers.

The exact focus of the PhD will therefore be determined along the way, based on input from the companies and your own interests. You will become a member of the EAVISE group in Sint‑Katelijne‑Waver, centrally located in Flanders and easily accessible by train or car.

• You have a master’s degree in computer science, AI, informatics, or a similar field.
• You have solid experience with language models and how to integrate them into applications.
• Experience with formal systems is not required, but is a plus.
• You speak and write English fluently. Knowledge of Dutch is a nice plus.
• You are interested in neuro‑symbolic AI systems and are motivated to make AI systems more reliable.
• You are able to collaborate smoothly with others.
• You will be offered a full‑time PhD position for 1 year, extendable to 4 years after a first positive evaluation.
• You will conduct research leading to a doctoral degree.
• You will become part of a dynamic research group, with access to computer infrastructure and travel budgets.
• You will have the opportunity to attend international conferences and build a scientific network.