Efstathia Soufleri

Efstathia Soufleri works in computer science on the intersection of machine learning privacy, robustness, and applications in areas like mental health, finance, and video understanding. Recent work spans differentially private classifiers for social media mental health analysis, benchmarking large language models for financial and multilingual tasks, and studying how deep learning models memorize or leak information through metrics like loss curvature. This suggests a research profile combining core machine learning methods (privacy, robustness, model compression) with applied projects in healthcare, finance, and social media analysis.

Publication output has grown notably in recent years, rising from little to no activity in the early-to-mid 2010s decade to a peak of several papers per year in 2024-2026.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026