Short Bio

Efstathia Soufleri is a Postdoctoral Fellow with the Archimedes Unit at the Athena Research Center, Greece. She earned her Ph.D. in Electrical and Computer Engineering from Purdue University, USA, where she specialized in Efficient and Privacy-Preserving Machine Learning Algorithms under the guidance of Prof. Kaushik Roy. 

Her work has been published in leading venues including ICML, NeurIPS, ICLR, EMNLP, TMLR, and DATE, with a Spotlight Presentation at NeurIPS 2024. She has also been recognized for her service to the research community as a Top Reviewer at NeurIPS 2025. Her contributions span privacy-preserving optimization, large-scale memorization analysis, multimodal language model evaluation, and efficient model design

She holds a Master of Science in Computer Science from the University of Thessaly, where she developed efficient video compression algorithms, and a Bachelor's degree in Applied Mathematics from the National and Kapodistrian University of Athens, which she completed in 3.5 years with honors

In recognition of her academic excellence, she has received several distinctions, including the Gerondelis Foundation Award for Academic Excellence, a scholarship from the Greek State Scholarships Foundation, and the distinction of being named Valedictorian at the University of Thessaly.

Research Interests

Her research focuses on privacy-preserving and secure machine learning, memorization and data leakage in large language models (LLMs), and the development of large-scale empirical frameworks for analyzing generalization, robustness, and trustworthiness in modern AI systems.


⚬ Recent publication: ICLR 2026 poster "Memorization Through the Lens of Sample Gradients"