PhD Studentship: Retrieval Augmented Generation (RAG) in the Charity Sector

Applications are invited for a PhD studentship in the Department of Computer Science. The successful candidate will have the opportunity to work on innovative techniques in Retrieval Augmented Generation (RAG) for Charity Sector projects.

We are seeking a PhD candidate to join our research group focused on improving large language models (LLMs) through retrieval-augmented generation (RAG). The successful candidate will work at the intersection of information retrieval, natural language processing, and machine learning. The position is fully funded by the school and a charity partner with the restriction that the funding is for a home student.

Research Focus would include (but not limited to):

• Developing novel architectures for efficient and accurate retrieval mechanisms in RAG systems.
• Investigating methods to improve context relevance and reduce hallucination in language models.
• Exploring dynamic knowledge integration techniques for real-time information updates.
• Creating evaluation frameworks for RAG system oriented at rigorously benchmarking the. performance and reliability across domains.

Eligibility and requirements 

The candidate should have a distinction or merit MSc (or equivalent, or higher) degree in Data Science, AI, or a closely related subject. They should demonstrate aptitude for original research.

Scholarship: The studentship is for 3 years and will provide an annual tax-free stipend that tracks the UKRI rate (currently £21,237 pa) and tuition fees (Home only). 

Additional income: Each student may also have the opportunity to earn around £2,200/year on an average (max. is around £4,300/year) through a teaching assistantship.

The candidate should possess strong programming skills (Python, PyTorch/TensorFlow), and have an excellent academic record and research potential. A candidate who demonstrates exceptional aptitude in one or more of these areas (as evidenced, for instance, through strong academic credentials or research papers in reputable, peer-reviewed journals/conferences) may be accorded preference. Ideally, the successful candidate should have proven skills in NLP, machine learning, and information retrieval as well as experience with transformer architectures and language models.

A doctoral candidate is expected to meet the following pre-requisites for their PhD: 

• Demonstrate a sound knowledge of their research area;
• Achieve and demonstrate significant depth in at least a few chosen sub-areas relevant to their primary research area;
• Demonstrate the ability to conduct independent research, including a critical assessment of their own and others’ research;

If you are interested in applying, I will be happy to address any initial informal enquiries: https://www.city.ac.uk/about/people/academics/chris-child

How to apply

Online applications should be submitted by clicking the 'Apply' button, above.

For queries regarding the application process, please contact pgr.sst.enquire@city.ac.uk

City, University of London is committed to promoting equality, diversity and inclusion in all its activities, processes, and culture for our whole community, including staff, students and visitors.

We welcome applications regardless of age, caring responsibilities, disability, gender identity, gender reassignment, marital status, nationality, pregnancy, race and ethnic origin, religion and belief, sex, sexual orientation and socio-economic background. City operates a guaranteed interview scheme for disabled applicants.

Advert information