Security & Privacy

Abstract Gradient Training: A Unified Certification Framework for Data Poisoning, Unlearning, and Differential Privacy JMLR 2026 Accelerating Confidential Databases with Crypto-Free Mappings OSDI 2026 Acumen: A Platform for Encrypted and Accountable Collaborative Editing OSDI 2026 Adaptive and Robust Watermark for Generative Tabular Data UAI 2026 Agent-based Substructure Counting under Local Differential Privacy ACL 2026 AudiFair: Privacy-Preserving Framework for Auditing Fairness UAI 2026 A Unified Perspective on Adversarial Membership Manipulation in Vision Models CVPR 2026 Black-Box Membership Inference Attacks for Video Training Data in Multimodal Large Language Models ACL 2026 Black-box Membership Inference Attacks on the Pre-training Data of Image-generation Models CVPR 2026 CheckMIABench: Firm Foundations For Membership Inference Attacks on Language Models ACL 2026 Classification Under Local Differential Privacy with Model Reversal and Model Averaging JMLR 2026 Cocoon: A System Architecture for Differentially Private Training with Correlated Noises OSDI 2026 Designing to Forget: Deep Semi-parametric Models for Unlearning CVPR 2026 Differentially Private 2D Human Pose Estimation CVPR 2026 Differentially Private Approval-Based Committee Voting UAI 2026 Differentially Private Best-Arm Identification JMLR 2026 Differentially Private Bilevel Optimization ALT 2026 Differentially Private Estimation and Inference in High-Dimensional Regression with FDR Control JMLR 2026 Differentially Private Language Generation and Identification in the Limit (Extended Abstract) COLT 2026 Do Vision-Language Models Leak What They Learn? Adaptive Token-Weighted Model Inversion Attacks CVPR 2026 FedAlign: Differentially Private Distribution Alignment for Non-IID Federated Learning CVPR 2026 FedMOP: Achieving Enhanced Privacy and Performance in Federated Learning via Momentum Orthogonal Projection CVPR 2026 Fixed-Parameter Tractability of Private Synthetic Data Generation COLT 2026 Frequency-domain Manipulation for Face Obfuscation CVPR 2026 From Measurement to Mitigation: Quantifying and Reducing Identity Leakage in Image Representation Encoders with Linear Subspace Removal CVPR 2026