3 papers
cs.AI2026
diffGHOST: Diffusion based Generative Hedged Oblivious Synthetic Trajectories
Florent Guépin, Cheick Tidiani Cisse, Denis Renaud +2
Trajectories are nowadays valuable information for a wide range of applications. However they are also inherently sensitive, as they contain highly personal information about indiv…
cs.AI2026
A Dual Perspective on Synthetic Trajectory Generators: Utility Framework and Privacy Vulnerabilities
Aya Cherigui, Florent Guépin, Arnaud Legendre +1
Human mobility data are used in numerous applications, ranging from public health to urban planning. Human mobility is inherently sensitive, as it can contain information such as r…
cs.LG2025
Lost in the Averages: A New Specific Setup to Evaluate Membership Inference Attacks Against Machine Learning Models
NataÅ¡a KrÄo, Florent Guépin, Matthieu Meeus +2
Synthetic data generators and machine learning models can memorize their training data, posing privacy concerns. Membership inference attacks (MIAs) are a standard method of estima…