一句话：提出基于多模态分析与多 LLM 智能体协作的 Android 欺诈应用检测工具，准确率达 91.7%。
针对虚假、恶意等欺诈性 Android 应用日益泛滥的问题，论文提出检测工具 AgentDroid。传统检测方法难以处理多模态数据且误报率高，AgentDroid 通过处理 Android 应用并抽取一组多模态数据进行分析，再由多个承担专门角色的 LLM 智能体对相关数据分别解读并协作研判，从而有效识别复杂欺诈行为。作者构建了包含多种欺诈类别与合法应用的数据集，并在该数据集上验证工具效果。实验结果显示，基于 GPT-4o 的多智能体工具取得 91.7% 的准确率和 91.68% 的 F1 值，优于所对比的基线方法。
With the increasing prevalence of fraudulent Android applications such as fake and malicious applications, it is crucial to detect them with high accuracy and adaptability. We present AgentDroid, a novel tool for Android fraudulent application detection based on multi-modal analysis and multi-agent systems. AgentDroid overcomes the limitations of traditional detection methods such as the inability to handle multimodal data and high false alarm rates. It processes Android applications and extracts a series of multi-modal data for analysis. Multiple LLM-based agents with specialized roles analyze the relevant data and collaborate to detect complex fraud effectively. We curated a dataset containing various categories of fraudulent applications and legitimate applications and validated our tool on this dataset. Experimental results indicate that our multi-agent tool based on GPT-4o achieves an accuracy of 91.7% and an F1-Score of 91.68%, outperforming the baseline methods. A video of AgentDroid is available at https://youtu.be/YOM9Ex-nBts.
DOI 原文 ·
@inproceedings{paperbot9464,
title = {AgentDroid: A Multi-Agent Tool for Detecting Fraudulent Android Applications},
author = {Ruwei Pan and Hongyu Zhang and Zhonghao Jiang and Ran Hou},
booktitle = {ASE},
year = {2025},
doi = {10.1109/ase63991.2025.00362}
}