paper
Paper accepted at ICML 2026
Our paper "OptiFluence: Principled Design of Privacy Canaries" has been accepted at ICML 2026. See you in Seoul!
PhD in Computer Science· University of Toronto | Vector Institute
I'm a first year PhD student in machine learning at the University of Toronto & Vector Institute supervised by Professor Nicolas Papernot.
My work focuses on understanding the training dynamics of large models to improve both capability and safety. I am interested in interpretability, mechanistic understanding, data attribution, and improving LLM's causal inference and reasoning abilities.
Previously, I studied Engineering Science (Machine Intelligence) at the University of Toronto.
News
Our paper "OptiFluence: Principled Design of Privacy Canaries" has been accepted at ICML 2026. See you in Seoul!