? PhD opportunity in Montréal — Human-Guided Evaluation of Generative AI Systems
How can we evaluate AI systems in realistic situations—and decide which tests or expert judgments will provide the most valuable evidence about their risks?
Together with Jackie Cheung and our partners at Waterloo, Université Aix-Marseille and Orange, I am seeking a PhD candidate within REGARD project to explore these questions through probabilistic modelling, adaptive testing, simulated users, and targeted human feedback.
The research will connect mathematical foundations with experiments on language models, focusing on reliability, uncertainty, and the productive use of human expertise.
? Montréal, with research activities at ILLS and Mila
? Intended start — January 2027
? Possibility of a cotutelle between Université Paris-Saclay and McGill University
? Competitive funding
? Rolling applications until the position is filled
We welcome applicants from computer science, mathematics, statistics, engineering, and related disciplines who enjoy combining theory with experimentation.
Full description and application details https://pablo-piantanida.org/openings/phd-ai-evaluation/