Security

ARGUS: Defending Against Multimodal Indirect Prompt Injection via Steering Instruction-Following Behavior CVPR 2026 On the Relevance of Byzantine Robust Optimization Against Data Poisoning JMLR 2026 PGA: Prior-free Generative Attack for Practical No-box Scenario CVPR 2026 Collapsing Sequence-Level Data-Policy Coverage via Poisoning Attack in Offline Reinforcement Learning UAI 2025 Cutting Through Privacy: A Hyperplane-Based Data Reconstruction Attack in Federated Learning UAI 2025 Influence-Based Fair Selection for Sample-Discriminative Backdoor Attack AAAI 2025 RepeatLeakage: Leak Prompts from Repeating as Large Language Model Is a Good Repeater AAAI 2025 SALSA: A Secure, Adaptive and Label-Agnostic Scalable Algorithm for Machine Unlearning UAI 2025 Not All Prompts Are Secure: A Switchable Backdoor Attack Against Pre-trained Vision Transfomers CVPR 2024 PKAD: Pretrained Knowledge is All You Need to Detect and Mitigate Textual Backdoor Attacks EMNLP 2024 Defending against Data-Free Model Extraction by Distributionally Robust Defensive Training NeurIPS 2023 Preprocessors Matter! Realistic Decision-Based Attacks on Machine Learning Systems ICML 2023 A Unified Evaluation of Textual Backdoor Learning: Frameworks and Benchmarks NeurIPS 2022 Robust learning under clean-label attack COLT 2021 SecureNets: Secure Inference of Deep Neural Networks on an Untrusted Cloud ACML 2018