PhD Studentship: Interpreting Language Representations in Large Language Models: A Computational Model for Measuring Linguistic Capabilities

Large Language Models (LLMs) are reshaping how we interact with language technologies, yet many questions remain about what these models actually “know” about language. The University of Exeter is offering one funded PhD position investigating how LLMs represent and process linguistic knowledge, drawing on tools from computer science, linguistics, and neuroscience. The aim is to build new computational models that help us interpret and control how language is generated and understood by AI systems.

The selected student will join the University’s vibrant research environment and will be encouraged to begin focused research early on. To complement their background, optional training in natural language processing, machine learning, and linguistics will be available. During the first year, the student will define their main research direction in collaboration with academic supervisors.

This funded PhD is suitable for applicants with a background in Natural Language Processing or Computational Linguistics, with a degree in Computer Science, Artificial Intelligence, Linguistics or Engineering. Applicants with strong programming skills and interests in language models, interpretability, and human-centred NLP are especially encouraged to apply.

For eligible students, the studentship will cover home or international tuition fees for 3.5 years full-time. The student will be based at the University of Exeter’s Streatham Campus. The studentship will be awarded on the basis of merit. International applicants are eligible to apply, but should note that the award may only partially cover the full international tuition fee. International applicants should to be aware that they will have to cover the cost of their student visa, healthcare surcharge and other costs of moving to the UK to do a PhD. They will also need to cover the cost of a student visa, healthcare surcharge, and relocation.

Advert information