Yves-Alexandre de Montjoye

Checkpoint-GCG: Auditing and Attacking Fine-Tuning-Based Prompt Injection Defenses

The Tail Tells All: Estimating Model-Level Membership Inference Vulnerability Without Reference Models

Unique between the lines: benchmarking re-identification risk for text anonymization

Certification for Differentially Private Prediction in Gradient-Based Training

Exploring the limits of strong membership inference attacks on large language models

Lost in the Averages: Evaluating record-specific MIAs against Machine Learning models