arXiv 2408.133662026-03-26cs.CLcs.AIcs.LG

CodeRefine: 一种提升研究论文中大语言模型生成代码实现的管道

CodeRefine: A Pipeline for Enhancing LLM-Generated Code Implementations of Research Papers

Ekaterina Trofimova, Emil Sataev, Abhijit Singh Jowhari

备注：: The results mentioned in the paper are non-reproducible. We have rechecked the metrics, and they do not match with the ones that have been provided in the paper. Therefore, we accept that this article is neither suitable nor up to the mark for the scientific community and must be with-drawn. We fully understand the consequences, and would like to wishfully retract this article

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

CodeRefine通过多步骤方法将论文方法转化为功能代码，利用预定义本体构建知识图谱，并通过回顾性检索增强生成方法提升代码准确性，有效解决理论研究与实践实现之间的桥梁问题。

摘要：

本文提出了CodeRefine，一种利用大语言模型（LLMs）将研究论文方法转化为功能代码的新型框架。我们的多步骤方法首先从论文中提取并总结关键文本片段，分析其代码相关性，并使用预定义本体构建知识图谱。然后从该结构化表示中生成代码，并通过提出的回顾性检索增强生成方法进行增强。CodeRefine解决了理论研究与实际实现之间的挑战，提供了一种比LLM零样本提示更准确的替代方案。在多样化的科学论文上的评估显示，CodeRefine能够提高论文中代码的实现质量，可能加速前沿算法在现实应用中的采用。

英文摘要：

This paper presents CodeRefine, a novel framework for automatically transforming research paper methodologies into functional code using Large Language Models (LLMs). Our multi-step approach first extracts and summarizes key text chunks from papers, analyzes their code relevance, and creates a knowledge graph using a predefined ontology. Code is then generated from this structured representation and enhanced through a proposed retrospective retrieval-augmented generation approach. CodeRefine addresses the challenge of bridging theoretical research and practical implementation, offering a more accurate alternative to LLM zero-shot prompting. Evaluations on diverse scientific papers demonstrate CodeRefine's ability to improve code implementation from the paper, potentially accelerating the adoption of cutting-edge algorithms in real-world applications.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。