揭示恶意软件检测中虚假相关性的实际影响，量化编译空隙的误导作用。
Empirical Quantification of Spurious Correlations in Malware Detection
- 通过分析编译器留下的空隙特征，发现模型依赖虚假相关性
- 在小规模数据集上验证了两种端到端模型对空隙的依赖程度差异
- 为生产部署提供可量化的模型选择依据，适合安全研究者参考
端到端深度学习在恶意软件检测中表现卓越，但其成功源于利用虚假相关性——即推理时看似重要、实则无意义的特征。尽管已有研究指出深度网络主要依赖元数据，却未进一步量化这些特征对决策的实际影响。本文通过对一个小型平衡数据集的开创性分析，揭示了模型对编译器留下的空隙的依赖程度，该现象削弱了编译后代码的实际相关性。在此基础上，我们对两种端到端模型进行了排名，以更清晰地判断哪种更适合投入实际生产环境。
原文摘要 · Abstract (English)
End-to-end deep learning exhibits unmatched performance for detecting malware, but such an achievement is reached by exploiting spurious correlations -- features with high relevance at inference time, but known to be useless through domain knowledge. While previous work highlighted that deep networks mainly focus on metadata, none investigated the phenomenon further, without quantifying their impact on the decision. In this work, we deepen our understanding of how spurious correlation affects deep learning for malware detection by highlighting how much models rely on empty spaces left by the compiler, which diminishes the relevance of the compiled code. Through our seminal analysis on a small-scale balanced dataset, we introduce a ranking of two end-to-end models to better understand which is more suitable to be put in production.
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。