Security

Phantom: Physical Object Interactions as Dynamic Triggers for NMS-Exploited Backdoors CVPR 2026 Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern CVPR 2026 PIArena: A Platform for Prompt Injection Evaluation ACL 2026 PoInit-of-View: Poisoning Initialization of Views Transfers Across Multiple 3D Reconstruction Systems CVPR 2026 Privacy-Preserving Robustness Verification for Neural Networks UAI 2026 PROMPRINT: Prompt Fingerprinting via First-Token Response for LLM App Cloning Detection ACL 2026 Protecting Language Models Against Unauthorized Distillation through Trace Rewriting ACL 2026 PureProof: Diffusion-Resistant Black-box Targeted Attack on Large Vision-Language Models CVPR 2026 QuantileMark: A Message-Symmetric Multi-bit Watermark for LLMs ACL 2026 R$^2$TUA: Reconstruction-residual Based Targeted and Untargeted Attack Against Text-Image Person Re-Identification CVPR 2026 RaPA: Enhancing Transferable Targeted Attacks via Random Parameter Pruning CVPR 2026 RAPIDS: Resume Attack Prompt Injection Detection at Scale ACL 2026 ReasMark: A Robust Watermark for Attributing LLM Reasoning Under Knowledge Distillation Attacks ACL 2026 ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha Moments ACL 2026 RecoverMark: Robust Watermarking for Localization and Recovery of Manipulated Faces CVPR 2026 RedCoder: Automated Multi-Turn Red Teaming for Code LLMs ACL 2026 Red-teaming Retrieval-Augmented Diffusion Models via Poisoning Knowledge Bases CVPR 2026 Reference Attack: A New Cross-Modal Jailbreaking Attack against Multimodal Large Language Models ACL 2026 Rel-Zero: Harnessing Patch-Pair Invariance for Robust Zero-Watermarking Against AI Editing CVPR 2026 RemedyGS: Defend 3D Gaussian Splatting Against Computation Cost Attacks CVPR 2026 Retrievals Can Be Detrimental: Unveiling the Backdoor Vulnerability of Retrieval-Augmented Diffusion Models ACL 2026 RevINN: An End-to-End Invertible Neural Network for Reversible Adversarial Examples Generation CVPR 2026 Robust3DGSW: Toward Robust Watermarking for Quantization-Aware 3D Gaussian Splatting CVPR 2026 Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion Models CVPR 2026 Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization ACL 2026