ARMOR++: Agentic Orchestration of a Multi-Domain Primitive Set for Transferable Attacks on Deepfake Detectors
ARMOR++:用于对深度伪造检测器进行可转移攻击的多域原语集的智能编排
机构 * Department of Informatics, Aristotle University of Thessaloniki(塞萨洛尼基亚里士多德大学信息学系) ; Infocomm Technology Cluster, Singapore Institute of Technology(新加坡科技学院信息通信技术集群) ; Department of Applied Physics and Applied Mathematics, Columbia University(哥伦比亚大学应用物理与应用数学系) ; Division of Natural and Applied Sciences, Duke Kunshan University(昆山杜克大学自然科学与应用科学部) ; Department of Electrical and Computer Engineering, University of Toronto(多伦多大学电气与计算机工程系)
AI总结 研究针对深度伪造检测器在黑盒对抗转移下可靠性下降问题,提出ARMOR++多智能体框架,利用视觉与语言模型提供语义先验并编排原语选择等,整合多种原语有效针对异构偏差,实验表明其性能优于现有基线,凸显当前检测器可靠性差距及智能编排的有效性。