Source Code
Source code, the fundamental building block of software, is the subject of intense research focusing on improving its analysis, generation, and security. Current efforts leverage machine learning, particularly transformer-based models like BERT and GPT variants, and graph neural networks, to analyze code for vulnerabilities, predict defects, and even automatically generate code from natural language descriptions. These advancements have significant implications for software development, enhancing code quality, security, and developer productivity, while also raising new challenges related to code authorship attribution and the detection of AI-generated code.
Papers
A Controlled Experiment on the Energy Efficiency of the Source Code Generated by Code Llama
Vlad-Andrei Cursaru, Laura Duits, Joel Milligan, Damla Ural, Berta Rodriguez Sanchez, Vincenzo Stoico, Ivano Malavolta
GraphSL: An Open-Source Library for Graph Source Localization Approaches and Benchmark Datasets
Junxiang Wang, Liang Zhao
Exploring the Impact of the Output Format on the Evaluation of Large Language Models for Code Translation
Marcos Macedo, Yuan Tian, Filipe R. Cogo, Bram Adams
CodeS: Natural Language to Code Repository via Multi-Layer Sketch
Daoguang Zan, Ailun Yu, Wei Liu, Dong Chen, Bo Shen, Wei Li, Yafen Yao, Yongshun Gong, Xiaolin Chen, Bei Guan, Zhiguang Yang, Yongji Wang, Qianxiang Wang, Lizhen Cui