Smart Contract Vulnerability Detection using Graph Neural Network

Authors: Yuan Zhuang, Zhenguang Liu, Peng Qian, Qi Liu, Xiang Wang, Qinming He

IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments show that our proposed approach significantly outperforms state-of-the-art methods in detecting three different types of vulnerabilities.
Researcher Affiliation Academia Yuan Zhuang1, , Zhenguang Liu1, , Peng Qian1, , Qi Liu2 , Xiang Wang3 , Qinming He4 1Zhejiang Gongshang University 2University of Oxford 3 National University of Singapore 4 Zhejiang University
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No Our implementations are released to facilitate future research. (No specific link provided in the paper itself)
Open Datasets No The paper mentions using "real-world smart contract datasets as ESC (Ethereum Smart Contracts) and VSC (VNT chain Smart Contracts)" collected from platforms like "http://etherscan.io/address/0xbb9bc244 d798123fde783fcc1c72d3bb8c189413" and "https://github.com/vntchain/go-vnt". However, it does not provide explicit access information (DOI, direct download link to the curated dataset used) for the *specific* datasets constructed and used in the experiments.
Dataset Splits No For each dataset, we randomly pick 20% contracts as the training set while the remainings are utilized for the testing set. (No explicit mention of a validation set split)
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments.
Software Dependencies No The paper does not specify the software dependencies with version numbers (e.g., programming language, libraries, frameworks) used for implementation or experiments.
Experiment Setup No The paper describes the model architecture and data processing steps but does not provide specific experimental setup details such as hyperparameter values (e.g., learning rate, batch size, number of epochs) or optimizer settings.