I’m an ML researcher working on generative modelling (especially diffusion and flow models) and probabilistic inference, with a particular emphasis on sampling. My recent work adapts probabilistic methods for post-training and test-time scaling in both continuous and discrete diffusion models, including applications to problems in physics and biology. This is a direction I’m keen to develop further.
I’m currently a PhD student at Mila, supervised by Yoshua Bengio. Previously, I completed my master’s thesis on Bayesian federated learning with Pascal Poupart.
Education
PhD, Computer Science — University of Montreal / Mila
Sept 2023 – expected May 2027
Advisor: Yoshua Bengio
Research Master's, Computer Science — University of Waterloo
Sept 2021 – Aug 2023
Advisor: Pascal Poupart
BASc, Electrical and Computer Engineering — University of Toronto
Sept 2016 – May 2021
Experience
Engineering Intern, Display Team — Qualcomm
May 2019 – Aug 2020
Designed image-processing algorithms for image quality enhancement, including a learning-based method for simulating high-dynamic-range images from standard-dynamic-range content.
Publications
My publications are also available on Google Scholar.
* denotes equal contribution.
Preprint · Under review, May 2026
Preprint · Under review, Jan 2026
ICML 2025
Approximate Inference Workshop, 2024