Not an A11y: How Android Accessibility Exposes Mobile AI Agents to Indirect Prompt Injection 文章

ArXiv CS.AI2026-08-11PAPERen作者: Rahul Deivasigamani, Sayeda Faatin Alvi, Derqui Andrea, Kaushal Punjabi, Stjepan Picek
查看原文 →
关系图谱

Not an A11y: How Android Accessibility Exposes Mobile AI Agents to Indirect Prompt Injection · 相关技术

相关技术

unauthorized actionscontext driftgoal hijackingadversarial promptsIndirect Prompt InjectionAndroid accessibility (A11y) treesautonomous agents