Large language models in software engineering
Nguyen Van Viet, Nguyen The Vinh
Abstract
Large Language Models (LLMs) are very large deep learning models pre-trained on a vast amount of data. This article aims to provide an overview of the use of major language models in the field of software engineering from January 2021 to February 2024. It surveys the emerging area of Large Language Modeling in Software Engineering but acknowledges that to fully understand the issues, effects, and limitations of LLMs in this field, further research is needed. The article also highlights open research challenges for applying Large Language Models to technical problems faced by software engineers. The exceptional properties of LLMs bring novelty and creativity to applications within Software Engineering activities, including coding, design, requirements, repair, refactoring, performance improvement, documentation, and analytics. Our survey demonstrates the key role of reliable and efficient large language models in the development and deployment of Software Engineering.