Generating Move Smart Contracts based on Concepts

TL;DR

ConMover is a framework that improves smart contract code generation in Move by combining a knowledge graph, few verified examples, and iterative refinement, significantly enhancing accuracy for low-resource scenarios.

Contribution

It introduces a novel approach integrating concept retrieval, planning, coding, and debugging to improve LLM-based Move smart contract generation.

Findings

Substantial accuracy improvements over baseline models

Effective in low-resource code generation scenarios

Bridges gap between natural language and reliable smart contracts

Abstract

The growing adoption of formal verification for smart contracts has spurred the development of new verifiable languages like Move. However, the limited availability of training data for these languages hinders effective code generation by large language models (LLMs). This paper presents ConMover, a novel framework that enhances LLM-based code generation for Move by leveraging a knowledge graph of Move concepts and a small set of verified code examples. ConMover integrates concept retrieval, planning, coding, and debugging agents in an iterative process to refine generated code. Evaluations with various open-source LLMs demonstrate substantial accuracy improvements over baseline models. These results underscore ConMover's potential to address low-resource code generation challenges, bridging the gap between natural language descriptions and reliable smart contract development.