Function-to-Style Guidance of LLMs for Code Translation
This addresses the problem of limited adoption of LLMs in real-world software development by enhancing code translation quality, though it appears incremental as it builds on existing LLM capabilities.
The paper tackled the challenge of ensuring both correctness and readability in code translation using LLMs, proposing F2STrans, which improved performance significantly, enabling a smaller model (Qwen-1.5B) to outperform larger models like Qwen-32B and GPT-4 across 20 scenarios.
Large language models (LLMs) have made significant strides in code translation tasks. However, ensuring both the correctness and readability of translated code remains a challenge, limiting their effective adoption in real-world software development. In this work, we propose F2STrans, a function-to-style guiding paradigm designed to progressively improve the performance of LLMs in code translation. Our approach comprises two key stages: (1) Functional learning, which optimizes translation correctness using high-quality source-target code pairs mined from online programming platforms, and (2) Style learning, which improves translation readability by incorporating both positive and negative style examples. Additionally, we introduce a novel code translation benchmark that includes up-to-date source code, extensive test cases, and manually annotated ground-truth translations, enabling comprehensive functional and stylistic evaluations. Experiments on both our new benchmark and existing datasets demonstrate that our approach significantly improves code translation performance. Notably, our approach enables Qwen-1.5B to outperform prompt-enhanced Qwen-32B and GPT-4 on average across 20 diverse code translation scenarios.