Privacy

Cocoon: A System Architecture for Differentially Private Training with Correlated Noises OSDI 2026 Differentially Private Best-Arm Identification JMLR 2026 Do Vision-Language Models Leak What They Learn? Adaptive Token-Weighted Model Inversion Attacks CVPR 2026 Improving the Convergence of Private Shuffled Gradient Methods with Public Data UAI 2026 Balls-and-Bins Sampling for DP-SGD AISTATS 2025 Person De-reidentification: A Variation-guided Identity Shift Modeling CVPR 2025 SALSA: A Secure, Adaptive and Label-Agnostic Scalable Algorithm for Machine Unlearning UAI 2025 The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD ICLR 2025 A Collocation-based Method for Addressing Challenges in Word-level Metric Differential Privacy ACL 2024 Center-Based Relaxed Learning Against Membership Inference Attacks UAI 2024 Disguise without Disruption: Utility-Preserving Face De-identification AAAI 2024 Private Language Models via Truncated Laplacian Mechanism EMNLP 2024 Protecting Privacy in Classifiers by Token Manipulation ACL 2024 Thinking Outside of the Differential Privacy Box: A Case Study in Text Privatization with Language Model Prompting EMNLP 2024 Disparate Impact in Differential Privacy from Gradient Misalignment ICLR 2023 CryptoGCN: Fast and Scalable Homomorphically Encrypted Graph Convolutional Network Inference NeurIPS 2022 Large Language Models Can Be Strong Differentially Private Learners ICLR 2022 Large-Scale Differentially Private BERT EMNLP 2022 Do not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning ICLR 2021 Differentially Private Language Models Benefit from Public Pre-training EMNLP 2020 ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition CVPR 2020