Leveraging Large Language Models for Code Translation and Software Development in Scientific Computing

2024-10-31

arXiv.org

AI Summary

This work addresses the challenge of modernizing legacy Fortran code in Large Hadron Collider (LHC) simulation software within high-energy physics. We propose a trustworthy, generative-AI–driven code migration methodology targeting Fortran-to-C++ translation, Fortran–C interoperability API generation, and structured code analysis. Our novel CodeScribe framework integrates domain-aware prompt engineering, static-analysis–guided API contract modeling, real-time developer supervision, and human-in-the-loop verification. The approach overcomes the trustworthiness bottleneck of AI-generated code in scientific computing: it achieves high-fidelity translation of core modules, produces production-ready interoperable interfaces, improves code refactoring efficiency by over 3×, and reduces error rates to levels acceptable for manual review. By enabling rigorous validation and collaborative refinement, CodeScribe establishes a verifiable, collaborative paradigm for modernizing scientific software heritage.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Search and Optimization: Sampling/Simulation-based Search
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

The emergence of foundational models and generative artificial intelligence (GenAI) is poised to transform productivity in scientific computing, especially in code development, refactoring, and translating from one programming language to another. However, because the output of GenAI cannot be guaranteed to be correct, manual intervention remains necessary. Some of this intervention can be automated through task-specific tools, alongside additional methodologies for correctness verification and effective prompt development. We explored the application of GenAI in assisting with code translation, language interoperability, and codebase inspection within a legacy Fortran codebase used to simulate particle interactions at the Large Hadron Collider (LHC). In the process, we developed a tool, CodeScribe, which combines prompt engineering with user supervision to establish an efficient process for code conversion. In this paper, we demonstrate how CodeScribe assists in converting Fortran code to C++, generating Fortran-C APIs for integrating legacy systems with modern C++ libraries, and providing developer support for code organization and algorithm implementation. We also address the challenges of AI-driven code translation and highlight its benefits for enhancing productivity in scientific computing workflows.

Problem

• Automating code translation between programming languages using GenAI
• Ensuring correctness in AI-generated code through verification methodologies
• Enhancing productivity in scientific computing workflows with AI tools

Innovation

• CodeScribe tool combines prompt engineering with supervision
• Automates Fortran to C++ code translation and API generation
• Enhances productivity in scientific computing workflows