Abstract
The rapid advancements in large language models (LLMs) have opened newavenues across various fields, including cybersecurity, which faces anever-evolving threat landscape and need for innovative technologies. Despiteinitial explorations into the application of LLMs in cybersecurity, there is alack of a comprehensive overview of this research area. This paper bridge thisgap by providing a systematic literature review, encompassing an analysis ofover 180 works, spanning across 25 LLMs and more than 10 downstream scenarios.Our comprehensive overview addresses three critical research questions: theconstruction of cybersecurity-oriented LLMs, LLMs' applications in variouscybersecurity tasks, and the existing challenges and further research in thisarea. This study aims to shed light on the extensive potential of LLMs inenhancing cybersecurity practices, and serve as a valuable resource forapplying LLMs in this doamin. We also maintain and regularly updated list ofpractical guides on LLMs for cybersecurity athttps://github.com/tmylla/Awesome-LLM4Cybersecurity.