Reza Shokri

Information-Theoretic Membership Inference for Granular Quantification of Memorization

LLMs Leak Training Data Beyond Verbatim Memorization via Membership Decoding

Generalization in LLM Problem Solving: The Case of the Shortest Path

How much of my dataset did you use? Quantitative Data Usage Inference in Machine Learning

Contextual Integrity in LLMs via Reasoning and Reinforcement Learning

The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text

Minerva: A Programmable Memory Test Benchmark for Language Models

The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text

Identifying Optimal Output Sets for Differential Privacy Auditing

Watermark Smoothing Attacks against Language Models