用基于XLNet的方法进行软件漏洞检测的远程上下文建模,使用基于XLNet的方法
Yinhu Zhao1, Guanjun Lin2,3, Zhenxuan Liao4,5
1School of Electronic, Electrical Engineering and Physics, Fujian University of Technology, Fuzhou, Fujian, 350118, China.
Scientific reports
|January 16, 2026
概括
XLNetVD通过使用XLNet来捕获长代码依赖性来增强软件漏洞检测,以68%的F1得分优于现有模型. 这个框架提供了一个最先进的解决方案,用于识别代码中的漏洞.
科学领域:
- 网络安全 网络安全
- 软件工程 软件工程 软件工程
- 人工智能的人工智能
背景情况:
- 软件漏洞检测对于网络安全至关重要.
- 基于语言模型 (LM) 的方法显示出希望,但与远程代码依赖性作斗争.
- 变压器架构在捕获广泛的代码上下文方面存在局限性.
研究的目的:
- 介绍XLNetVD,这是一个基于XLNet的框架,用于功能级别的漏洞检测.
- 解决现有模型在捕获远程代码依赖性方面的局限性.
- 评估XLNet在漏洞检测方面的有效性.
主要方法:
- 利用双向变压器-XL模型进行扩展的上下文建模.
- 与六个上下文和三个非上下文嵌入模型对比的XLNet.
- 将XLNet集成到一个端到端的框架中,XLNetVD,并应用低级调整 (LoRA) 微调.
主要成果:
- XLNet获得了最高的F1得分68%,超过了CodeBERT和GPT-2.
- 在LoRA增强的LM中,XLNet-LoRA显示出最佳的性能效率权衡.
- 在不平衡的真实世界和平衡的SARD数据集上,XLNetVD表现出了竞争力.
结论:
- XLNetVD 确立了自己作为软件漏洞检测的最先进的解决方案.
- 该框架有效地捕捉了必要的代码依赖性,用于识别微妙的漏洞.
- 基于XLNet的方法比现有的基于LM的方法提供了显著的改进.
相关概念视频
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...