Code Summarization
Code summarization aims to automatically generate concise natural language descriptions of source code, improving code understanding and maintainability. Current research heavily utilizes large language models (LLMs), often within encoder-decoder architectures or enhanced with techniques like prompt engineering, retrieval-augmented mechanisms, and multi-task learning, to improve summary quality and address challenges like handling diverse programming languages and code structures. This field is significant because effective code summarization can significantly reduce the time and effort required for software development, maintenance, and comprehension, impacting both research and practical software engineering workflows.
Papers
What Do They Capture? -- A Structural Analysis of Pre-Trained Language Models for Source Code
Yao Wan, Wei Zhao, Hongyu Zhang, Yulei Sui, Guandong Xu, Hai Jin
Source Code Summarization with Structural Relative Position Guided Transformer
Zi Gong, Cuiyun Gao, Yasheng Wang, Wenchao Gu, Yun Peng, Zenglin Xu