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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

One of the most pressing threats to computing systems is softwarevulnerabilities, which can compromise both hardware and software components.Existing methods for vulnerability detection remain suboptimal. Traditionaltechniques are both time-consuming and labor-intensive, whilemachine-learning-based approaches often underperform when applied to complexdatasets, due to their inability to capture high-dimensional relationships.Previous deep-learning strategies also fall short in capturing sufficientfeature information. Although self-attention mechanisms can process informationover long distances, they fail to capture structural information. In thispaper, we introduce DefectHunter, an innovative model for vulnerabilityidentification that employs the Conformer mechanism. This mechanism fusesself-attention with convolutional networks to capture both local, position-wisefeatures and global, content-based interactions. Furthermore, we optimize theself-attention mechanisms to mitigate the issue of excessive attention headsintroducing extraneous noise by adjusting the denominator. We evaluatedDefectHunter against ten baseline methods using six industrial and two highlycomplex datasets. On the QEMU dataset, DefectHunter exhibited a 20.62\%improvement in accuracy over Pongo-70B, and for the CWE-754 dataset, itsaccuracy was 14.64\% higher. To investigate how DefectHunter comprehendsvulnerabilities, we conducted a case study, which revealed that our modeleffectively understands the mechanisms underlying vulnerabilities.

DefectHunter: A Novel LLM-Driven Boosted-Conformer-based Code Vulnerability Detection Mechanism