Security

RunawayEvil: Jailbreaking the Image-to-Video Generative Models CVPR 2026 Safe-FedLLM: Delving into the Safety of Federated Large Language Models ACL 2026 SAIDO: Generalizable Detection of AI-Generated Images via Scene-Aware and Importance-Guided Dynamic Optimization in Continual Learning CVPR 2026 SecureVibeBench: Benchmarking Secure Vibe Coding of AI Agents via Reconstructing Vulnerability-Introducing Scenarios ACL 2026 Securing the Tool Layer: A Threat Taxonomy and Runtime Defense Framework for Model Context Protocol Deployments ACL 2026 Selective Knowledge Distillation: Fusing LLM Semantic Strengths with DNN Efficiency for Binary Code Similarity Detection ACL 2026 SharedRequest: Privacy-Preserving Model-Agnostic Inference for Large Language Models ACL 2026 SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models CVPR 2026 SoundBreak: A Systematic Study of Audio-Only Adversarial Attacks on Trimodal Models ACL 2026 SSG: Logit-Balanced Vocabulary Partitioning for LLM Watermarking ACL 2026 Stealing Split Learning Bottom Models by Recovering Embedding Geometry CVPR 2026 Steganography Beyond Pixels: Reimagining Image Steganography as Cross-Modal Linguistic Communication ACL 2026 Stop Fixating on Prompts: Reasoning Hijacking and Constraint Tightening for Red-Teaming LLM Agents ACL 2026 Synthetic Text Detection in the Age of Large Language Models: Watermark vs. Automatic Detection ACL 2026 Test-Time Attention Purification for Backdoored Large Vision Language Models CVPR 2026 The Coherence Trap: When MLLM-Crafted Narratives Exploit Manipulated Visual Contexts CVPR 2026 Toward Secure Tuning: Mitigating Security Risks from Instruction Fine-Tuning ACL 2026 Towards Stealthy and Effective Backdoor Attacks on Lane Detection: A Naturalistic Data Poisoning Approach CVPR 2026 TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards ACL 2026 UMMF: Protecting Copyright of Large Vision-Language Models through Unlearning-based Multimodal Memorization Fingerprint ACL 2026 Unleashing Stealthy Backdoor Pandemic by Infecting a Single Diffusion Model CVPR 2026 V-Attack: Targeting Disentangled Value Features for Controllable Adversarial Attacks on LVLMs CVPR 2026 VCP-Attack: Visual-Contrastive Projection for Transferable Black-Box Targeted Attacks on Large Vision-Language Models CVPR 2026 Verifiable LLM-Generated Text Detection via Projected Semantic-Structural Distributions ACL 2026 VerilogLAVD: LLM-Aided Pattern Generation for Verilog CWE Detection ACL 2026