PhD student, University of Toronto
3 papers at NeurIPS 2025
We use influence functions to attribute and suppress training examples that promote toxic behaviors in LLMs.
We scale the influence-function-based data valuation method to recent LLMs and their massive training datasets.
We apply the EKFAC-preconditioner on Neumann series iterations to arrive at an unbiased iHVP approximation for TDA that improves influence function and unrolled differentiation performance.
