PhD Candidate in Human-Guided Evaluation of Generative AI Systems

The REGARD project seeks a PhD candidate to investigate how to evaluate AI systems in realistic settings by developing methods that integrate probabilistic modelling, adaptive testing, simulated users, and targeted human feedback. The research will bridge mathematical foundations with experiments on language models, emphasizing reliability, uncertainty quantification, and the efficient use of human expertise. The position is based in Montréal, with research activities at ILLS and Mila, and offers the possibility of a cotutelle between Université Paris-Saclay and McGill University. Competitive funding is available, and applications are reviewed on a rolling basis until the position is filled.