Coca: Improving and Explaining Graph Neural Network-Based Vulnerability Detection Systems. (arXiv:2401.14886v1 [cs.CR])

Recently, Graph Neural Network (GNN)-based vulnerability detection systems
have achieved remarkable success. However, the lack of explainability poses a
critical challenge to deploy black-box models in security-related domains. For
this reason, several approaches have been proposed to explain the decision
logic of the detection model by providing a set of crucial statements
positively contributing to its predictions. Unfortunately, due to the
weakly-robust detection models and suboptimal explanation strategy, they have
the danger of revealing spurious correlations and redundancy issue.

In …

arxiv box challenge coca critical decision deploy detection domains graph logic network neural network security systems vulnerability vulnerability detection