Software Vulnerability Detection Using a Lightweight Graph Neural Network

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Farmer, Miles, Ufuktepe, Ekincan, Watson, Anne, Carvalho, Hialo Muniz, Okun, Vadim, Maasaoui, Zineb, Palaniappan, Kannappan
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917372428812288
author Farmer, Miles
Ufuktepe, Ekincan
Watson, Anne
Carvalho, Hialo Muniz
Okun, Vadim
Maasaoui, Zineb
Palaniappan, Kannappan
author_facet Farmer, Miles
Ufuktepe, Ekincan
Watson, Anne
Carvalho, Hialo Muniz
Okun, Vadim
Maasaoui, Zineb
Palaniappan, Kannappan
contents Large Language Models (LLMs) have emerged as a popular choice in vulnerability detection studies given their foundational capabilities, open source availability, and variety of models, but have limited scalability due to extensive compute requirements. Using the natural graph relational structure of code, we show that our proposed graph neural network (GNN) based deep learning model VulGNN for vulnerability detection can achieve performance almost on par with LLMs, but is 100 times smaller in size and fast to retrain and customize. We describe the VulGNN architecture, ablation studies on components, learning rates, and generalizability to different code datasets. As a lightweight model for vulnerability analysis, VulGNN is efficient and deployable at the edge as part of real-world software development pipelines.
format Preprint
id arxiv_https___arxiv_org_abs_2603_29216
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Software Vulnerability Detection Using a Lightweight Graph Neural Network
Farmer, Miles
Ufuktepe, Ekincan
Watson, Anne
Carvalho, Hialo Muniz
Okun, Vadim
Maasaoui, Zineb
Palaniappan, Kannappan
Software Engineering
Artificial Intelligence
Cryptography and Security
Machine Learning
Large Language Models (LLMs) have emerged as a popular choice in vulnerability detection studies given their foundational capabilities, open source availability, and variety of models, but have limited scalability due to extensive compute requirements. Using the natural graph relational structure of code, we show that our proposed graph neural network (GNN) based deep learning model VulGNN for vulnerability detection can achieve performance almost on par with LLMs, but is 100 times smaller in size and fast to retrain and customize. We describe the VulGNN architecture, ablation studies on components, learning rates, and generalizability to different code datasets. As a lightweight model for vulnerability analysis, VulGNN is efficient and deployable at the edge as part of real-world software development pipelines.
title Software Vulnerability Detection Using a Lightweight Graph Neural Network
topic Software Engineering
Artificial Intelligence
Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2603.29216