arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2408.07526cs.SEcs.CRcs.LG
收藏

面向漏洞检测的基于学习的模型:一项广泛研究

Learning-based Models for Vulnerability Detection: An Extensive Study

School of Software Technology, Zhejiang University(浙江大学软件学院);Zhejiang University(浙江大学);Central University of Finance and Economics(中央财经大学)

Chao Ni, Liyu Shen, Xiaodan Xu, Xin Yin, Shaohua Wang

AI总结:

该研究针对现有漏洞检测深度学习模型认知不足的问题,从五个维度对两类主流基于学习的漏洞检测方法开展大规模实验调研,明确了不同方法的性能差异、特性及实际应用要点。

摘要:

尽管许多基于深度学习的模型已在漏洞检测领域取得显著进展,但我们对这些模型的认知仍较为匮乏,这限制了模型能力的进一步提升、对模型检测机制的理解,以及模型实际应用的效率与安全性。本文通过在一个近期构建的大规模数据集上开展实验,对两类最先进的基于学习的方法(基于序列的方法和基于图的方法)进行了广泛且全面的调研。我们从五个维度探究了七个研究问题,分别为模型能力、模型可解释性、模型稳定性、模型易用性以及模型经济性。我们通过实验证明了基于序列的模型的优势,以及LLM(ChatGPT)和基于图的模型的能力局限性。我们探究了基于学习的模型擅长检测的漏洞类型,并揭示了即使输入发生语义等价的细微改变,模型仍存在不稳定性。我们通过实证解释了模型究竟学习到了什么。我们总结了模型的预处理流程以及易用性要求。最后,我们初步归纳了这些模型实现经济、安全的实际应用所需的关键信息。

英文摘要:

Though many deep learning-based models have made great progress in vulnerability detection, we have no good understanding of these models, which limits the further advancement of model capability, understanding of the mechanism of model detection, and efficiency and safety of practical application of models. In this paper, we extensively and comprehensively investigate two types of state-of-the-art learning-based approaches (sequence-based and graph-based) by conducting experiments on a recently built large-scale dataset. We investigate seven research questions from five dimensions, namely model capabilities, model interpretation, model stability, ease of use of model, and model economy. We experimentally demonstrate the priority of sequence-based models and the limited abilities of both LLM (ChatGPT) and graph-based models. We explore the types of vulnerability that learning-based models skilled in and reveal the instability of the models though the input is subtlely semantical-equivalently changed. We empirically explain what the models have learned. We summarize the pre-processing as well as requirements for easily using the models. Finally, we initially induce the vital information for economically and safely practical usage of these models.

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。

↑