Nicolas Papernot

Breach By A Thousand Leaks: Unsafe Information Leakage in 'Safe' AI Responses ICLR 2025 Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention ICML 2025 Fast Exact Unlearning for In-Context Learning Data for LLMs ICML 2025 Language Models May Verbatim Complete Text They Were Not Explicitly Trained On ICML 2025 Leveraging Per-Instance Privacy for Machine Unlearning ICML 2025 Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settings ICML 2025 Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model ICLR 2025 Auditing Private Prediction ICML 2024 Confidential-DPproof: Confidential Proof of Differentially Private Training ICLR 2024 LLM Dataset Inference: Did you train on my dataset? NeurIPS 2024 Memorization in Self-Supervised Learning Improves Downstream Generalization ICLR 2024 Position: Fundamental Limitations of LLM Censorship Necessitate New Approaches ICML 2024 Temporal-Difference Learning Using Distributed Error Signals NeurIPS 2024 The Fundamental Limits of Least-Privilege Learning ICML 2024 Architectural Backdoors in Neural Networks CVPR 2023 Confidential-PROFITT: Confidential PROof of FaIr Training of Trees ICLR 2023 Flocks of Stochastic Parrots: Differentially Private Prompt Learning for Large Language Models NeurIPS 2023 Have it your way: Individualized Privacy Assignment for DP-SGD NeurIPS 2023 Measuring Forgetting of Memorized Training Examples ICLR 2023 Robust and Actively Secure Serverless Collaborative Learning NeurIPS 2023 Training Private Models That Know What They Don’t Know NeurIPS 2023 A Zest of LIME: Towards Architecture-Independent Model Distances ICLR 2022 Dataset Inference for Self-Supervised Models NeurIPS 2022 Hyperparameter Tuning with Renyi Differential Privacy ICLR 2022 Increasing the Cost of Model Extraction with Calibrated Proof of Work ICLR 2022 In Differential Privacy, There is Truth: on Vote-Histogram Leakage in Ensemble Private Learning NeurIPS 2022 Is Fairness Only Metric Deep? Evaluating and Addressing Subgroup Gaps in Deep Metric Learning ICLR 2022 On the Difficulty of Defending Self-Supervised Learning against Model Extraction ICML 2022 On the Limitations of Stochastic Pre-processing Defenses NeurIPS 2022 The Privacy Onion Effect: Memorization is Relative NeurIPS 2022 Washing The Unwashable : On The (Im)possibility of Fairwashing Detection NeurIPS 2022 CaPC Learning: Confidential and Private Collaborative Learning ICLR 2021 Data-Free Model Extraction CVPR 2021 Dataset Inference: Ownership Resolution in Machine Learning ICLR 2021 Label-Only Membership Inference Attacks ICML 2021 Manipulating SGD with Data Ordering Attacks NeurIPS 2021 Markpainting: Adversarial Machine Learning meets Inpainting ICML 2021 Tempered Sigmoid Activations for Deep Learning with Differential Privacy AAAI 2021 Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial Perturbations ICML 2020 Thieves on Sesame Street! Model Extraction of BERT-based APIs ICLR 2020 Analyzing and Improving Representations with the Soft Nearest Neighbor Loss ICML 2019 MixMatch: A Holistic Approach to Semi-Supervised Learning NeurIPS 2019 Adversarial Examples that Fool both Computer Vision and Time-Limited Humans NeurIPS 2018 Ensemble Adversarial Training: Attacks and Defenses ICLR 2018 Scalable Private Learning with PATE ICLR 2018