14606 - Research Associate

The School of Informatics, University of Edinburgh invites applications for a 2-year Post-Doctoral Research Associate (PDRA), to do research on reliable constrained generation with large language models via neuro-symbolic models, with applications to biological data, under the supervision of Dr Antonio Vergari and in collaboration with Prof. Wouter Boomsma from the University of Copenhagen.

The Opportunity:

Don’t just see the bigger picture, help create it.

The position is in collaboration with Prof. Wouter Boomsma from the University of Copenhagen. As part of this project, we aim to investigate the theoretical and practical foundations of reliable controlled generation with large language models (LLMs). We will do so by applying principled probabilistic neuro-symbolic techniques and advance the current state-of-the-art in terms of reliability and efficiency. This ambitious goal will be tested on sequential biological data.

The PDRA will be part of the April Lab at the School of Informatics, University of Edinburgh which is ranked among the top schools in Europe for AI research according to CS Rankings. The PDRA will be supervised by Dr. Antonio Vergari, a leader in tractable probabilistic machine learning and neuro-symbolic AI, and will collaborate with researchers from Prof. Boomsma’s lab.

The PDRA role involves
1) conducting cutting-edge research in LLM constrained generation with neuro-symbolic layers, building on our lab’s pioneering research on reliable and trustworthy ML; and
2) assisting Prof. Boomsma’s team with applications to biological data;
3) writing scientific papers documenting the proposed methodology and presenting them at conferences.

This position includes funding for international travel to attend conferences and offers access to our HPC infrastructure. The position is open to UK and international applicants, with visa sponsorship available. This post is advertised as full-time (35 hours per week), however, we are open to considering part-time or flexible working patterns. We are also open to considering requests for hybrid working (on a non-contractual basis) that combines a mix of remote and regular on-campus working.