{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T22:26:37Z","timestamp":1790288797077,"version":"4.1.0"},"reference-count":24,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T00:00:00Z","timestamp":1779321600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Software vulnerability analysis is critical for maintaining secure and reliable systems, yet traditional Deep Learning (DL) models often act as \u201cblack boxes,\u201d lacking transparency and failing to leverage the explicit structural semantics of code. In this paper, we propose KG-HiAttention, a novel neuro-symbolic framework that synergizes sub-symbolic deep learning with symbolic AI-based Knowledge Graphs (KGs). We construct a CPG-inspired lightweight program graph for each software function, approximating control-flow (CFG) and data-flow (DFG) dependencies through line-level edges. This symbolic structure is processed by a Graph Attention Network (GAT) and fused with semantic embeddings from a pre-trained CodeT5 encoder through multimodal fusion (concatenation and MLP classifier). Experiments on the real-world BigVul dataset show that KG-HiAttention achieves competitive performance (AUC-ROC 0.763 \u00b1 0.009, five seeds), statistically equivalent to a strong Hybrid Ensemble baseline, while improving specificity from 0.321 (baseline) to 0.458 and providing graph-based explainability that the baseline cannot offer.<\/jats:p>","DOI":"10.3389\/frai.2026.1794125","type":"journal-article","created":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T05:44:40Z","timestamp":1779342280000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["KG-HiAttention: synergizing AI-based knowledge graphs and deep learning for explainable software vulnerability analysis"],"prefix":"10.3389","volume":"9","author":[{"given":"Francisco","family":"Pinto-Santos","sequence":"first","affiliation":[{"name":"BISITE Research Group, University of Salamanca","place":["Salamanca, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carolina","family":"Zato","sequence":"additional","affiliation":[{"name":"BISITE Research Group, University of Salamanca","place":["Salamanca, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"H\u00e9ctor","family":"Quinti\u00e1n","sequence":"additional","affiliation":[{"name":"Investigaci\u00f3n en TIC (CITIC), Universidade da Coru\u00f1a","place":["A Coru\u00f1a, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tian Cheng","family":"Li","sequence":"additional","affiliation":[{"name":"Key Lab, Information Fusion Tech, Northwestern Polytechnical University","place":["Xi'an, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pablo","family":"Chamoso","sequence":"additional","affiliation":[{"name":"BISITE Research Group, University of Salamanca","place":["Salamanca, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3212695","article-title":"A survey of machine learning for big code and naturalness","volume":"51","author":"Allamanis","year":"2018","journal-title":"ACM Comput. Surv"},{"key":"B2","first-page":"654","article-title":"\u201cDiversevul: a new vulnerable source code dataset for deep learning based vulnerability detection,\u201d","volume-title":"Proceedings of the 26th International Symposium on Research in Attacks, Intrusions and Defenses, RAID '23","author":"Chen","year":"2023"},{"key":"B3","doi-asserted-by":"publisher","first-page":"107982","DOI":"10.1016\/j.infsof.2025.107982","article-title":"Vdmpagr: a vulnerability detection model based on pointer analysis and graph representation","volume":"191","author":"Dong","year":"2026","journal-title":"Inf. Softw. Technol"},{"key":"B4","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1145\/3379597.3387501","article-title":"\u201cA c\/c++ code vulnerability dataset with code changes and cve summaries,\u201d","volume-title":"Proceedings of the 17th International Conference on Mining Software Repositories, MSR '20","author":"Fan","year":"2020"},{"key":"B5","doi-asserted-by":"crossref","first-page":"1536","DOI":"10.18653\/v1\/2020.findings-emnlp.139","article-title":"\u201cCodebert: a pre-trained model for programming and natural languages,\u201d","volume-title":"Findings of the Association for Computational Linguistics: EMNLP 2020","author":"Feng","year":"2020"},{"key":"B6","doi-asserted-by":"crossref","first-page":"608","DOI":"10.1145\/3524842.3528452","article-title":"\u201cLinevul: a transformer-based line-level vulnerability prediction,\u201d","volume-title":"Proceedings of the 19th International Conference on Mining Software Repositories, MSR '22","author":"Fu","year":"2022"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1109\/ICSE.2012.6227135","article-title":"\u201cOn the naturalness of software,\u201d","author":"Hindle","year":"2012","journal-title":"2012 34th International Conference on Software Engineering (ICSE)"},{"key":"B8","first-page":"518","article-title":"\u201cCc2vec: distributed representations of code changes,\u201d","volume-title":"Proceedings of the ACM\/IEEE 42nd International Conference on Software Engineering, ICSE'20","author":"Hoang","year":"2020"},{"key":"B9","doi-asserted-by":"publisher","first-page":"107978","DOI":"10.1016\/j.infsof.2025.107978","article-title":"Representation learning for coincidental correctness in fault localization","volume":"191","author":"Hu","year":"2026","journal-title":"Inf. Softw. Technol"},{"key":"B10","first-page":"521","article-title":"\u201cCost estimation model using fifth generation language technique for software maintenance project,\u201d","volume-title":"Proceedings of International Conference on Recent Innovations in Computing","author":"Islam","year":"2022"},{"key":"B11","doi-asserted-by":"publisher","first-page":"494","DOI":"10.1109\/TNNLS.2021.3070843","article-title":"A survey on knowledge graphs: representation, acquisition, and applications","volume":"33","author":"Ji","year":"2022","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B12","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2017","journal-title":"arXiv preprint arXiv:1609.02907"},{"key":"B13","first-page":"4768","article-title":"\u201cA unified approach to interpreting model predictions,\u201d","volume-title":"Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS'17","author":"Lundberg","year":"2017"},{"key":"B14","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1109\/TSE.2022.3144348","article-title":"Deeplinedp: towards a deep learning approach for line-level defect prediction","volume":"49","author":"Pornprasit","year":"2023","journal-title":"IEEE Trans. Softw. Eng"},{"key":"B15","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1007\/s10458-019-09408-y","article-title":"Explainability in human-agent systems","volume":"33","author":"Rosenfeld","year":"2019","journal-title":"Auton. Agents Multi-Agent Syst"},{"key":"B16","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","article-title":"The graph neural network model","volume":"20","author":"Scarselli","year":"2009","journal-title":"IEEE Trans. Neural Netw"},{"key":"B17","first-page":"6000","article-title":"\u201cAttention is all you need,\u201d","volume-title":"Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS'17","author":"Vaswani","year":"2017"},{"key":"B18","article-title":"Graph attention networks","author":"Velickovi\u0107","year":"2018","journal-title":"arXiv preprint arXiv:1710.10903"},{"key":"B19","doi-asserted-by":"crossref","first-page":"8696","DOI":"10.18653\/v1\/2021.emnlp-main.685","article-title":"\u201cCodet5: identifier-aware unified pre-trained encoder-decoder models for code understanding and generation,\u201d","volume-title":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","author":"Wang","year":"2021"},{"key":"B20","doi-asserted-by":"publisher","first-page":"107977","DOI":"10.1016\/j.infsof.2025.107977","article-title":"Evaluating and improving LLM-based competitive program generation","volume":"191","author":"Wei","year":"2026","journal-title":"Inf. Softw. Technol"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2014.44","article-title":"\u201cModeling and discovering vulnerabilities with code property graphs,\u201d","author":"Yamaguchi","year":"2014","journal-title":"2014 IEEE Symposium on Security and Privacy"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3505243","article-title":"A survey on deep learning for software engineering","volume":"54","author":"Yang","year":"2020","journal-title":"ACM Comput. Surv"},{"key":"B23","article-title":"\u201cDevign: effective vulnerability identification by learning comprehensive program semantics via graph neural networks,\u201d","author":"Zhou","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B24","doi-asserted-by":"publisher","first-page":"107993","DOI":"10.1016\/j.infsof.2025.107993","article-title":"Line-level defect prediction based on preceding line-aware and inter-line semantics enhancement","volume":"191","author":"Zhu","year":"2026","journal-title":"Inf. Softw. Technol"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1794125\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T05:44:43Z","timestamp":1779342283000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1794125\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,21]]},"references-count":24,"alternative-id":["10.3389\/frai.2026.1794125"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1794125","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,21]]},"article-number":"1794125"}}