{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T14:42:20Z","timestamp":1774449740788,"version":"3.50.1"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":16,"publisher":"Elsevier BV","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Detecting vulnerabilities in smart contracts is challenging due to their complex semantics, structural diversity, and class imbalance. Existing deep learning approaches often treat contracts as plain text, overlooking the rich structural information in Abstract Syntax Trees (ASTs). To address these limitations, we propose SmartContractTransformer-GAN (SCT-GAN), a multi-task transformer-based framework for vulnerability detection and adversarial contract generation. SCT-GAN introduces three key innovations: (1) fusion of source code tokens and AST paths for enhanced semantic and structural modeling, (2) hierarchical detection at contract and line levels, enabling fine-grained identification even with limited context, and (3) a syntax-aware GAN generator-discriminator loop producing realistic and semantically meaningful smart contract code. Evaluation shows that while contract-level detection does not significantly improve the state of the art, line-level detection benefits most from the generative components, leveraging richer latent representations to capture logical patterns in individual lines. SCT-GAN\u2019s modular, memory-efficient design supports large-scale auditing, continual adaptation to emerging vulnerabilities, and synthetic dataset generation for low-data scenarios. Overall, SCT-GAN provides a scalable, interpretable, and generative solution for proactive smart contract security, advancing automated auditing and vulnerability synthesis in blockchain ecosystems.<\/jats:p>","DOI":"10.2139\/ssrn.6468371","type":"posted-content","created":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T13:39:49Z","timestamp":1774445989000},"source":"Crossref","is-referenced-by-count":0,"title":["SmartContractTransformer-GAN (SCT-GAN): A Framework for Smart Contract Vulnerability Detection and Synthetic Generation"],"prefix":"10.2139","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-3749-9488","authenticated-orcid":true,"given":"Jo\u00e3o","family":"Cris\u00f3stomo","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0834-0275","authenticated-orcid":true,"given":"Fernando","family":"Bacao","sequence":"additional","affiliation":[]},{"given":"Victor","family":"Lobo","sequence":"additional","affiliation":[]}],"member":"78","reference":[{"key":"ref1","year":"2025","journal-title":"Category deep-dive: $2.2 billion was stolen in crypto-related hacks in 2024"},{"key":"ref2","article-title":"Machine learning methods for detecting smart contracts vulnerabilities within Ethereum blockchain-A review","author":"J Crisostomo","year":"2025","journal-title":"Expert Systems With Applications"},{"key":"ref3","author":"T Liu","journal-title":"The Role of Transformer Models in Advancing Blockchain Technology: A Systematic Survey"},{"key":"ref4","author":"Z Wei","year":"2024","journal-title":"A Knowledge Distillation-Enhanced Framework for Automated Smart Contract Auditing Using Fine-Tuned LLMs"},{"key":"ref5","author":"O Zaazaa","journal-title":"SmartLLMSentry: A Comprehensive LLM Based Smart Contract Vulnerability Detection Framework"},{"key":"ref6","author":"Z Feng","year":"2020","journal-title":"CodeBERT: A pre-trained model for programming and natural languages"},{"key":"ref7","doi-asserted-by":"crossref","first-page":"21431","DOI":"10.1109\/JIOT.2023.3294496","article-title":"An efficient smart contract vulnerability detector based on semantic contract graphs using approximate graph matching","volume":"10","author":"Y Zhang","year":"2023","journal-title":"IEEE Internet of Things Journal"},{"key":"ref8","author":"J Kim","year":"2024","journal-title":"Robust vulnerability detection in solidity-based ethereum smart contracts using fine-tuned transformer encoder models"},{"key":"ref9","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-025-99267-3","article-title":"Enhancing smart contract security using a code representation and GAN based methodology","volume":"15","author":"D K Murala","year":"2025","journal-title":"Scientific Reports"},{"key":"ref10","author":"N K Jha","journal-title":"ReLU's revival: On the entropic overload in normalization-free large language models"},{"key":"ref11","author":"K Irie","year":"2019","journal-title":"Language modeling with deep transformers"},{"key":"ref12","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1109\/MSP.2017.2765202","article-title":"Generative adversarial networks: An overview","volume":"35","author":"A Creswell","year":"2018","journal-title":"IEEE Signal Process Mag"},{"key":"ref13","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1016\/j.gltp.2022.04.020","article-title":"A review: Data pre-processing and data augmentation techniques","volume":"3","author":"K Maharana","year":"2022","journal-title":"Global Transitions Proceedings"},{"key":"ref14","doi-asserted-by":"crossref","first-page":"97515","DOI":"10.1109\/ACCESS.2024.3427829","article-title":"Compiler-guided generation network for smart contract data augmentation","volume":"12","author":"S-J Hwang","year":"2024","journal-title":"IEEE Access"},{"key":"ref15","journal-title":"Github [Internet"},{"key":"ref16","journal-title":"Smart Contract Vulnerabilities on the Ethereum Blockchain: A Current Perspective"}],"container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T13:43:21Z","timestamp":1774446201000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=6468371"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":16,"URL":"https:\/\/doi.org\/10.2139\/ssrn.6468371","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}