Publications
* denotes equal contribution.
Reliability
Reliability and safety of LLM-based evaluation systems.
AI Privacy
Machine unlearning, membership inference, and data reconstruction.
NeurIPS 2025
Rethinking Exact Unlearning under Exposure: Extracting Forgotten Data under Exact Unlearning in Large Language Models
Theory and Practice of Differential Privacy 2025
Winning the MIDST Challenge: New Membership Inference Attacks on Diffusion Models for Tabular Data Synthesis
NeurIPS 2024 Workshop on Behavioral ML
A New Approach to Generate Individual Level Data of Walled Garden Platforms: Linear Programming Reconstruction
Network Security (Path Validation)
Secure and efficient multipath validation algorithms.
