arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.04526cs.SEcs.AIcs.CL
收藏

推动与代码相关的语言模型

Advancing Language Models for Code-related Tasks

School of Computer Software, Tianjin University(天津大学软件学院)

Zhao Tian

AI总结:

本研究通过改进代码数据质量、模型架构和推理能力,推动语言模型在软件开发中的实际应用。

摘要:

近年来,语言模型(LMs)的进步在各种软件工程任务中推动了显著进展。然而,现有的LMs仍然在复杂的编程场景中遇到困难,这主要是由于数据质量、模型架构和推理能力的限制。本研究通过三个互补的方向系统性地解决这些挑战:(1)通过代码差异引导的对抗增强技术(CODA)和代码去噪技术(CodeDenoise)改进代码数据质量;(2)通过语法引导的代码LMs(LEAM和LEAM++)增强模型架构;(3)通过提示技术(muFiX)和基于代理的技术(Specine)推进模型推理。这些技术旨在促进LMs在软件开发中的实际应用,并进一步推进智能软件工程。

英文摘要:

Recent advances in language models (LMs) have driven significant progress in various software engineering tasks. However, existing LMs still struggle with complex programming scenarios due to limitations in data quality, model architecture, and reasoning capability. This research systematically addresses these challenges through three complementary directions: (1) improving code data quality with a code difference-guided adversarial augmentation technique (CODA) and a code denoising technique (CodeDenoise); (2) enhancing model architecture via syntax-guided code LMs (LEAM and LEAM++); and (3) advancing model reasoning with a prompting technique (muFiX) and an agent-based technique (Specine). These techniques aim to promote the practical adoption of LMs in software development and further advance intelligent software engineering.

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。

↑