Deep learning for code generation: a survey

作     者：Huangzhao ZHANG Kechi ZHANG Zhuo LI Jia LI Jia LI Yongmin LI Yunfei ZHAO Yuqi ZHU Fang LIU Ge LI Zhi JIN 

作者机构：Key Lab of High Confidence Software Technologies (Peking University) Ministry of Education School of Computer Science Peking University School of Computer Science and Engineering Beihang University 

出 版 物：《Science China(Information Sciences)》 (中国科学:信息科学(英文版))

年 卷 期：2024年第67卷第9期

页      面：5-40页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by National Natural Science Foundation of China (Grant Nos. 62192733, 62192731, 61751210, 62072007, 61832009, 62192730) supported by National Natural Science Foundation of China (Grant No. 62302021) 

主　　题：code generation automated software engineering deep learning large model artificial intelligence 

摘      要：In the past decade, thanks to the powerfulness of deep-learning techniques, we have witnessed a whole new era of automated code generation. To sort out developments, we have conducted a comprehensive review of solutions to deep learning-based code generation. In this survey, we generally formalize the pipeline and procedure of code generation and categorize existing solutions according to taxonomy from perspectives of architecture, model-agnostic enhancing strategy, metrics, and tasks. In addition, we outline the challenges faced by current dominant large models and list several plausible directions for future research. We hope that this survey may provide handy guidance to understanding, utilizing, and developing deep learning-based code-generation techniques for researchers and practitioners.