English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Current machine-learning based software vulnerability detection methods areprimarily conducted at the function-level. However, a key limitation of thesemethods is that they do not indicate the specific lines of code contributing tovulnerabilities. This limits the ability of developers to efficiently inspectand interpret the predictions from a learnt model, which is crucial forintegrating machine-learning based tools into the software developmentworkflow. Graph-based models have shown promising performance in function-levelvulnerability detection, but their capability for statement-level vulnerabilitydetection has not been extensively explored. While interpreting function-levelpredictions through explainable AI is one promising direction, we hereinconsider the statement-level software vulnerability detection task from a fullysupervised learning perspective. We propose a novel deep learning framework,LineVD, which formulates statement-level vulnerability detection as a nodeclassification task. LineVD leverages control and data dependencies betweenstatements using graph neural networks, and a transformer-based model to encodethe raw source code tokens. In particular, by addressing the conflictingoutputs between function-level and statement-level information, LineVDsignificantly improve the prediction performance without vulnerability statusfor function code. We have conducted extensive experiments against alarge-scale collection of real-world C/C++ vulnerabilities obtained frommultiple real-world projects, and demonstrate an increase of 105\% in F1-scoreover the current state-of-the-art.

LineVD: Statement-level Vulnerability Detection using Graph Neural Networks