Privacy

Deduplicating Training Data Mitigates Privacy Risks in Language Models ICML 2022 DeepAuth: A DNN Authentication Framework by Model-Unique and Fragile Signature Embedding AAAI 2022 Design and Verification of the Arm Confidential Compute Architecture OSDI 2022 Distillation-Resistant Watermarking for Model Protection in NLP EMNLP 2022 Efficient Federated Learning on Knowledge Graphs via Privacy-preserving Relation Embedding Aggregation EMNLP 2022 Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attacks EMNLP 2022 Faster Privacy Accounting via Evolving Discretization ICML 2022 Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations CVPR 2022 Fishing for User Data in Large-Batch Federated Learning via Gradient Magnification ICML 2022 FLVoogd: Robust And Privacy Preserving Federated Learning ACML 2022 Friendly Noise against Adversarial Noise: A Powerful Defense against Data Poisoning Attack NeurIPS 2022 GSmooth: Certified Robustness against Semantic Transformations via Generalized Randomized Smoothing ICML 2022 How reparametrization trick broke differentially-private text representation learning ACL 2022 Image Disentanglement Autoencoder for Steganography Without Embedding CVPR 2022 Improved Certified Defenses against Data Poisoning with (Deterministic) Finite Aggregation ICML 2022 Invernet: An Inversion Attack Framework to Infer Fine-Tuning Datasets through Word Embeddings EMNLP 2022 Justitia: Software Multi-Tenancy in Hardware Kernel-Bypass Networks NSDI 2022 LAMP: Extracting Text from Gradients with Language Model Priors NeurIPS 2022 Learning to Generate Inversion-Resistant Model Explanations NeurIPS 2022 Leveraging Adversarial Examples To Quantify Membership Information Leakage CVPR 2022 Marksman Backdoor: Backdoor Attacks with Arbitrary Target Class NeurIPS 2022 Measuring Data Reconstruction Defenses in Collaborative Inference Systems NeurIPS 2022 MExMI: Pool-based Active Model Extraction Crossover Membership Inference NeurIPS 2022 NEUROCRYPT: Coercion-Resistant Implicit Memory Authentication (Student Abstract) AAAI 2022 Neurotoxin: Durable Backdoors in Federated Learning ICML 2022