UniASM:无需微调的二进制代码相似性检测
UniASM: Binary Code Similarity Detection without Fine-tuning
AI总结:
本文通过消融实验分析二进制代码嵌入性能因素,提出富语义函数表示技术和基于 UniLM 的 UniASM 模型,在跨编译器、跨优化级别、跨混淆及漏洞搜索任务上显著优于现有方法。
摘要:
二进制代码相似性检测(BCSD)被广泛应用于漏洞搜索、恶意软件检测、克隆检测和补丁分析等各类二进制分析任务。近期研究表明,基于学习的二进制代码嵌入模型表现优于传统的基于特征的方法。然而,以往研究并未深入探究影响模型性能的关键因素。本文设计了大量消融实验来探索这些影响因素,实验结果为我们带来了许多新的认识。我们在代码表示和模型选择两方面均进行了创新:提出了一种新颖的富语义函数表示技术,以确保模型能够捕捉二进制代码的复杂细微差别;并引入了首个基于 UniLM 的二进制代码嵌入模型,命名为 UniASM,该模型包含两个新设计的训练任务,用于学习二进制函数的表示。实验结果表明,UniASM 在评估数据集上的表现优于最先进(SOTA)方法。与最佳基线方法相比,其在跨编译器、跨优化级别和跨混淆场景下的 Recall@1 平均得分分别提升了 12.7%、8.5% 和 22.3%。此外,在已知漏洞搜索这一真实世界任务中,UniASM 的表现也优于当前所有基线。
英文摘要:
Binary code similarity detection (BCSD) is widely used in various binary analysis tasks such as vulnerability search, malware detection, clone detection, and patch analysis. Recent studies have shown that the learning-based binary code embedding models perform better than the traditional feature-based approaches. However, previous studies have not delved deeply into the key factors that affect model performance. In this paper, we design extensive ablation studies to explore these influencing factors. The experimental results have provided us with many new insights. We have made innovations in both code representation and model selection: we propose a novel rich-semantic function representation technique to ensure the model captures the intricate nuances of binary code, and we introduce the first UniLM-based binary code embedding model, named UniASM, which includes two newly designed training tasks to learn representations of binary functions. The experimental results show that UniASM outperforms the state-of-the-art (SOTA) approaches on the evaluation datasets. The average scores of Recall@1 on cross-compilers, cross-optimization-levels, and cross-obfuscations have improved by 12.7%, 8.5%, and 22.3%, respectively, compared to the best of the baseline methods. Besides, in the real-world task of known vulnerability search, UniASM outperforms all the current baselines.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。