News

Jun 01, 2026 I am joining the Vector Institute in Toronto as a Visiting PhD Student with Nicolas Papernot in the Clever Hans Lab. :canada:
May 15, 2026 Our new preprint, Assessing Per-Sample Membership Inference Vulnerability without Retraining, is now live on arXiv. The paper presents an efficient approach to evaluate per-sample membership inference attack (MIA) vulnerability while bypassing the overhead of retraining shadow models. Check it out here.
Oct 01, 2025 I started my PhD in the DATA Team at ENS-PSL, working on Differential Privacy and Membership Inference Attacks, under the supervision of Olivier Cappé and Jamal Atif. :mortar_board: