会员服务
DefenSee: Dissecting Threat from Sight and Text - A Multi-View Defensive Pipeline for Multi-modal Jailbreaks
Arxiv
0+阅读 · 12月1日
Immunity memory-based jailbreak detection: multi-agent adaptive guard for large language models
Arxiv
0+阅读 · 12月3日
Jailbreaking in the Haystack
Arxiv
0+阅读 · 11月5日
"To Survive, I Must Defect": Jailbreaking LLMs via the Game-Theory Scenarios
Arxiv
0+阅读 · 11月20日
Jailbreaking and Mitigation of Vulnerabilities in Large Language Models
Arxiv
0+阅读 · 11月25日
Practical and Stealthy Touch-Guided Jailbreak Attacks on Deployed Mobile Vision-Language Agents
Arxiv
0+阅读 · 11月20日
Defending Large Language Models Against Jailbreak Exploits with Responsible AI Considerations
Arxiv
0+阅读 · 11月24日
参考链接
微信扫码咨询专知VIP会员
Top