The Fine Print of Misbehavior: VRP’s Blueprint and Safety Stance Post date August 11, 2025 Post author By Large Models (dot tech) Post categories In adversarial-ai-research, ai-evaluation, ai-model-security, ethical-ai-attacks, mllm-jailbreak, role-play-attack, text-moderation, vrp-methodology
One Image to Rule Them All: The Jailbreak That Outsmarts Multimodal AI Post date August 11, 2025 Post author By Large Models (dot tech) Post categories In adversarial-ai, ai-alignment-bypass, ai-model-security, future-ai-research, mllm-jailbreak, role-play-attack, universal-jailbreak, visual-role-play
Introducing VRP: Structure-Based Role-Play Attacks on Multimodal Large Language Models Post date August 11, 2025 Post author By Large Models (dot tech) Post categories In adversarial-ai, ai-misuse-prevention, ai-model-security, mllm-jailbreak, multimodal-ai, role-play-attack, universal-jailbreak, visual-role-play
Sciencx
Sciencx