Marius Irinel Banu
Research Subjects
Marius Irinel Banu has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Geanina Silviana Banu
Department of Informatics, Faculty of Informatics, Titu Maiorescu University, 040051 Bucharest, Romania
1 shared publication - Iustin Priescu
Department of Informatics, Faculty of Informatics, Titu Maiorescu University, 040051 Bucharest, Romania
1 shared publication - Tatiana Corina Dosescu
“Titu Maiorescu” University , Bucharest , Romania
1 shared publication
Latest Publications
- Open Access
- 1 June 2025
- Published by Walter de Gruyter GmbH in Land Forces Academy Review
- Vol. 30 (2) , 291-302
Prompt Engineering in Cybersecurity – Achieving Technological Edge
Prompt Engineering has emerged as a critical technique for optimizing interactions with Large Language Models (LLMs), enhancing their precision and effectiveness in various applications. This paper explores the role of structured prompting techniques in cybersecurity, focusing on their potential to improve threat detection, security automation, and incident response. By leveraging Chain-of-Thought (CoT) and Multimodal CoT prompting, the study evaluates how well-crafted prompts enhance LLM capabilities in analyzing network traffic, identifying phishing attempts, automating security reporting, and conducting forensic investigations. The findings demonstrate that structured prompts significantly improve LLM performance, enabling more accurate anomaly detection, faster security incident analysis, and enhanced cyber threat intelligence. Additionally, the study outlines a framework for integrating Prompt Engineering into cybersecurity workflows, illustrating its potential for AI-driven security monitoring and defense strategies. These insights contribute to the ongoing advancement of AI applications in cybersecurity, highlighting the transformative role of LLMs and Prompt Engineering in strengthening digital security measures.
- 1 June 2025
- Published by Walter de Gruyter GmbH in Land Forces Academy Review
- Vol. 30 (2) , 291-302
Open AccessPrompt Engineering in Cybersecurity – Achieving Technological Edge
Prompt Engineering has emerged as a critical technique for optimizing interactions with Large Language Models (LLMs), enhancing their precision and effectiveness in various applications. This paper explores the role of structured prompting techniques in cybersecurity, focusing on their potential to improve threat detection, security automation, and incident response. By leveraging Chain-of-Thought (CoT) and Multimodal CoT prompting, the study evaluates how well-crafted prompts enhance LLM capabilities in analyzing network traffic, identifying phishing attempts, automating security reporting, and conducting forensic investigations. The findings demonstrate that structured prompts significantly improve LLM performance, enabling more accurate anomaly detection, faster security incident analysis, and enhanced cyber threat intelligence. Additionally, the study outlines a framework for integrating Prompt Engineering into cybersecurity workflows, illustrating its potential for AI-driven security monitoring and defense strategies. These insights contribute to the ongoing advancement of AI applications in cybersecurity, highlighting the transformative role of LLMs and Prompt Engineering in strengthening digital security measures.