用多模态智能体高效识别手机广告，精准率提升超30%。
MANA: Towards Efficient Mobile Ad Detection via Multimodal Agentic UI Navigation
- 构建多模态智能体，结合静态、视觉与时间信号导航界面。
- 检测准确率提升30.5%-56.3%，探索步骤减少29.7%-63.3%。
- 适合移动安全审计，可发现隐蔽恶意广告。
移动端广告主导应用变现，但带来体验侵扰乃至恶意软件风险。现有方法或依赖静态分析（忽略运行时行为），或采用启发式界面探索（难应对稀疏与混淆广告）。本文提出MANA，首个用于移动端广告检测的智能体多模态推理框架。MANA将静态、视觉、时间与用户体验信号融合，生成有推理指导的界面导航策略，不仅决定如何遍历界面，还明确关注重点，实现高效稳健探索。我们在200个商用应用上部署并评估MANA，性能达当前最优。相比基线，准确率提升30.5%-56.3%，探索步数减少29.7%-63.3%。案例研究进一步验证其发现隐蔽及恶意广告的能力，凸显其在移动端广告审计中的实用性，并具拓展至运行时界面分析（如权限滥用）的潜力。代码与数据集见https://github.com/MANA-2026/MANA。
原文摘要 · Abstract (English)
Mobile advertising dominates app monetization but introduces risks ranging from intrusive user experience to malware delivery. Existing detection methods rely either on static analysis, which misses runtime behaviors, or on heuristic UI exploration, which struggles with sparse and obfuscated ads. In this paper, we present MANA, the first agentic multimodal reasoning framework for mobile ad detection. MANA integrates static, visual, temporal, and experiential signals into a reasoning-guided navigation strategy that determines not only how to traverse interfaces but also where to focus, enabling efficient and robust exploration. We implement and evaluate MANA on commercial smartphones over 200 apps, achieving state-of-the-art accuracy and efficiency. Compared to baselines, it improves detection accuracy by 30.5%-56.3% and reduces exploration steps by 29.7%-63.3%. Case studies further demonstrate its ability to uncover obfuscated and malicious ads, underscoring its practicality for mobile ad auditing and its potential for broader runtime UI analysis (e.g., permission abuse). Code and dataset are available at https://github.com/MANA-2026/MANA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。