基于异构图神经网络的智能合约漏洞检测与修复系统
The smart contract vulnerability detection and repair system based on heterogeneous graph neural networks solves the problems of low detection accuracy, high false alarm rate and insufficient automatic repair capability in existing technologies. It achieves efficient smart contract vulnerability detection and repair, improves detection accuracy and repair success rate, and adapts to new attack patterns.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG UNIV OF TECH
- Filing Date
- 2025-11-11
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies have low accuracy and high false alarm rate in smart contract vulnerability detection, cannot identify complex interactive vulnerabilities, lack automatic repair capabilities, and are poorly adaptable to new attack patterns.
A smart contract vulnerability detection and repair system based on heterogeneous graph neural networks is adopted. Through contract parsing, multi-layer graph construction, heterogeneous graph neural network module, vulnerability feature library, vulnerability identification engine and automatic repair module, combined with static taint analysis, formal verification and graph convolutional neural network, deep feature information extraction and automatic repair are achieved.
It achieves a significant improvement in the accuracy of smart contract vulnerability detection, a reduction in false positive rate, the ability to identify multi-level security risks, a high success rate of automatic repair, strong adaptability, and supports large-scale contract detection and repair.
Smart Images
Figure CN121543093B_ABST