Search papers, labs, and topics across Lattice.
SCPatcher leverages retrieval-augmented generation with a knowledge graph of 5,000 Ethereum contracts to automatically repair smart contract vulnerabilities. The knowledge graph captures function-level relationships, improving LLM reasoning for precise patching. SCPatcher achieves an 81.5% repair rate and 91.0% compilation pass rate, outperforming existing methods on a diverse set of vulnerable contracts.
Automated repair of smart contract vulnerabilities is now significantly more effective, achieving an 81.5% repair rate by combining LLMs with a knowledge graph of Ethereum contracts.
Smart contract vulnerabilities can cause substantial financial losses due to the immutability of code after deployment. While existing tools detect vulnerabilities, they cannot effectively repair them. In this paper, we propose SCPatcher, a framework that combines retrieval-augmented generation with a knowledge graph for automated smart contract repair. We construct a knowledge graph from 5,000 verified Ethereum contracts, extracting function-level relationships to build a semantic network. This graph serves as an external knowledge base that enhances Large Language Model reasoning and enables precise vulnerability patching. We introduce a two-stage repair strategy, initial knowledge-guided repair followed by Chain-of-Thought reasoning for complex vulnerabilities. Evaluated on a diverse set of vulnerable contracts, SCPatcher achieves 81.5\% overall repair rate and 91.0\% compilation pass rate, substantially outperforming existing methods.
