Research Fellow in NLP and Text-layout LLMs

The University of Surrey is a global community of ideas and people, dedicated to life-changing education and research. 

We have a bold vision of what we want to achieve - shaping ourselves into one of the best universities in the world, which we are achieving through the talents and endeavour of every employee.  

Our culture empowers people to achieve this aim and to collectively make a real difference.  

The role

Applications are invited for a 12-month Research Fellow in natural language processing (NLP) and text-layout large language models (LLMs), to work full-time on an ESRC-funded project “MetaCurate-ML: Extraction and Utilisation of Metadata from Non-machine-actionable Documents to Improve Data Curation and Discovery”. We are looking for the successful post holder to start as soon as possible and will be based in the Department of Computer Science and its highly regarded Nature-inspired Computing and Engineering (NICE) research group, reporting to the project PI (Dr. Suparna De).

The researcher will work on AI model training and inference, graph networks, and open standards/metadata. They will work within an ESRC project (‘MetaCurate-ML’) that aims to develop machine learning (ML) pipelines for processing of longitudinal pdf surveys into enhanced metadata following established metadata standards that support the FAIR principles. This is an excellent opportunity for a Research Fellow to further develop their research and software skills in an ambitious multi-disciplinary collaboration with social scientists, metadata curators  and data management specialists from CLOSER, University College London (UCL), the UK Data Service (UKDS) and Scotcen. 

The post-holder will be responsible for:

• Designing and developing ML models for the extraction of metadata from pdf survey questionnaires, taking advantage of recent developments in deep learning techniques, including multi-modal analysis, fusing textual and visual information (text-layout large language models (LLMs)), and zero-shot techniques, into a graph representation of the questionnaire elements. The developed models with take into account relevant open metadata standards in RDF/XML and related technologies. 
• Working collaboratively with software developers from CLOSER, UCL to develop relevant training datasets for the ML models.   

About you

• The successful applicant is expected to have a PhD (gained or near completion) or equivalent professional experience in computer science or a related subject.
• Experience in designing and developing ML models and state-of-the-art deep learning techniques.
• Experience in software development and creation of web-based tools and systems is desirable.
• Interdisciplinary research experience, working with researchers from data curation, social science, or health, will also be desirable.

For informal inquiries about the position, please contact Dr. Suparna De (s.de@surrey.ac.uk).

To apply

Please attached you CV and complete the questions on the application form evidencing your suitability for the role.

Interview Dates

Interviews will be held remotely over MS Teams on the 29th or 31st July 2024.

Further details:   

For more information and to apply online, please download the further details and click on the 'apply online' button above.

Advert information