Nicolas Papernot

Private Rate-Constrained Optimization with Applications to Fair Learning

OptiFluence: Scalable and Principled Design of Privacy Canaries

Data Forging Attacks on Cryptographic Model Certification

Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settings

Language Models May Verbatim Complete Text They Were Not Explicitly Trained On

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention

Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model

What Does It Take to Build a Performant Selective Classifier?

Breach By A Thousand Leaks: Unsafe Information Leakage in 'Safe' AI Responses

Leveraging Per-Instance Privacy for Machine Unlearning

Fast Exact Unlearning for In-Context Learning Data for LLMs

Lossgate: Incomplete Information and Misaligned Incentives Hinder Regulation of Societal Risks in Machine Learning