Harnessing Large Language Models for Real-Time Cyber Threat Detection and Response: A Comprehensive Survey
Year: 2025
Type: conference-paper
Abstract: Large Language Models (LLMs) have recently gained recognition as transformative tools across various domains, especially in cybersecurity. This survey explores the state-of-the-art applications of LLMs in tackling complex and rapidly evolving cyber threats. By synthesizing insights from recent literature, we analyze how LLMs enable real-time threat detection, detailed incident analysis, and actionable mitigation strategies. Compared to traditional cybersecurity approaches, LLMs demonstrate notable advancements in detection accuracy and response efficiency. We offer a comprehensive evaluation of the capabilities and limitations of LLM-based methodologies, spotlighting real-world use cases where these models have shown exceptional effectiveness. Additionally, we identify critical research challenges and provide future directions to enhance the performance, interpretability, and safety of LLM-driven cybersecurity systems.
Authors: Li, Xiaoyu, Sun, Nan, Jiang, Jiaojiao
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7117346418
FWCI:
Cites:
Cited by:
Related to: