SCALE: Constructing Structured Natural Language Comment Trees for Software Vulnerability Detection

arXiv:2403.19096v1 Announce Type: cross
Abstract: Recently, there has been a growing interest in automatic software vulnerability detection. Pre-trained model-based approaches have demonstrated superior performance than other Deep Learning (DL)-based approaches in detecting vulnerabilities. However, the existing pre-trained model-based approaches generally employ code sequences as input during prediction, and may ignore vulnerability-related structural information, as reflected in the following two aspects. First, they tend to fail to infer the semantics of the code statements with complex logic such as those containing …

arxiv automatic code cs.cr cs.se deep learning detection input interest language may natural natural language performance prediction scale software software vulnerability trees vulnerabilities vulnerability vulnerability detection