Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 9, 2026

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

KG-HiAttention: synergizing AI-based knowledge graphs and deep learning for explainable software vulnerability

Francisco Pinto-Santos1, Carolina Zato1, Héctor Quintián2

  • 1BISITE Research Group, University of Salamanca, Salamanca, Spain.

Frontiers in Artificial Intelligence
|June 8, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

KATENA: a verifiable governance architecture for encrypted cloud storage systems.

Frontiers in big data·2026
Same author

Beyond performance metrics: evaluating the unique value of generative AI in hybrid cybersecurity threat detection.

Frontiers in big data·2026
Same author

Deep Learning-Based 3D Reconstruction for Defect Detection in Shipbuilding Sub-Assemblies.

Sensors (Basel, Switzerland)·2026
Same author

A Fault-Detection System Approach for the Optimization of Warship Equipment Replacement Parts Based on Operation Parameters.

Sensors (Basel, Switzerland)·2023
Same author

A device and an app for the diagnosis and self-management of tinnitus.

Journal of integrative bioinformatics·2022
Same author

Taxi services and the carsharing alternative: a case study of valencia city.

Mathematical biosciences and engineering : MBE·2022
Same journal

Cross-linguistic patterns of cognitive biases in large language models: a comparative study in English, Hebrew, and Russian.

Frontiers in artificial intelligence·2026
Same journal

From human-like AI to user adoption: the role of trust, attitude, and social influence in shaping behavioral intention.

Frontiers in artificial intelligence·2026
Same journal

Building large-scale English-Romanian literary translation resources with open models.

Frontiers in artificial intelligence·2026
Same journal

Editorial: GenAI in healthcare: technologies, applications and evaluation.

Frontiers in artificial intelligence·2026
Same journal

Logic, inference, understanding: cross-domain generalization for generative language models.

Frontiers in artificial intelligence·2026
Same journal

Label tree semantic losses for rich multi-class medical image segmentation.

Frontiers in artificial intelligence·2026
See all related articles
This summary is machine-generated.

This study introduces KG-HiAttention, a neuro-symbolic framework for software vulnerability analysis. It enhances model transparency and explainability by combining deep learning with knowledge graphs.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Software Engineering

Background:

  • Traditional Deep Learning (DL) models for software vulnerability analysis lack transparency and fail to utilize code's structural semantics.
  • Existing methods often function as

Purpose of the Study:

  • To propose KG-HiAttention, a novel neuro-symbolic framework for improved software vulnerability analysis.
  • To enhance transparency and explainability in DL models for code security.

Main Methods:

  • Constructing a CPG-inspired lightweight program graph approximating control-flow (CFG) and data-flow (DFG) dependencies.
  • Processing the symbolic graph with a Graph Attention Network (GAT).
  • Fusing GAT outputs with CodeT5 semantic embeddings via multimodal fusion.
Keywords:
AI-based knowledge graphsCodeT5code property graphexplainable AI (XAI)graph attention networksneuro-symbolic AIsoftware vulnerability analysis

Related Experiment Videos

Last Updated: Jun 9, 2026

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

Main Results:

  • KG-HiAttention achieved competitive performance (AUC-ROC 0.763 ± 0.009) on the BigVul dataset.
  • Demonstrated statistically equivalent performance to a Hybrid Ensemble baseline.
  • Significantly improved specificity from 0.321 to 0.458.

Conclusions:

  • KG-HiAttention offers a transparent and explainable approach to software vulnerability analysis.
  • The neuro-symbolic framework effectively integrates structural code semantics with deep learning.
  • Provides valuable graph-based explainability absent in traditional black-box models.