Privacy

Mitigating Privacy Seesaw in Large Language Models: Augmented Privacy Neuron Editing via Activation Patching ACL 2024 Mixing Gradients in Neural Networks as a Strategy To Enhance Privacy in Federated Learning WACV 2024 Model Inversion Robustness: Can Transfer Learning Help? CVPR 2024 MultiLeg: Dataset for Text Sanitisation in Less-resourced Languages COLING 2024 NCIS: Neural Contextual Iterative Smoothing for Purifying Adversarial Perturbations WACV 2024 Neural Codec-based Adversarial Sample Detection for Speaker Verification Interspeech 2024 Noisy Neighbors: Efficient membership inference attacks against LLMs ACL 2024 On provable privacy vulnerabilities of graph representations NeurIPS 2024 On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift NeurIPS 2024 Open Problem: What is the Complexity of Joint Differential Privacy in Linear Contextual Bandits? COLT 2024 Optimal Locally Private Nonparametric Classification with Public Data JMLR 2024 Order of Magnitude Speedups for LLM Membership Inference EMNLP 2024 OSLO: One-Shot Label-Only Membership Inference Attacks NeurIPS 2024 PANORAMIA: Privacy Auditing of Machine Learning Models without Retraining NeurIPS 2024 PII-Compass: Guiding LLM training data extraction prompts towards the target PII via grounding ACL 2024 PostMark: A Robust Blackbox Watermark for Large Language Models EMNLP 2024 PPIDSG: A Privacy-Preserving Image Distribution Sharing Scheme with GAN in Federated Learning AAAI 2024 Practical Privacy-Preserving MLaaS: When Compressive Sensing Meets Generative Networks AAAI 2024 Preset-Voice Matching for Privacy Regulated Speech-to-Speech Translation Systems ACL 2024 Privacy Amplification by Iteration for ADMM with (Strongly) Convex Objective Functions AAAI 2024 Privacy Amplification for Matrix Mechanisms ICLR 2024 Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models NeurIPS 2024 Privacy-Constrained Policies via Mutual Information Regularized Policy Gradients AISTATS 2024 Privacy-Preserving Face Recognition Using Trainable Feature Subtraction CVPR 2024 Privacy-Preserving Optics for Enhancing Protection in Face De-Identification CVPR 2024