Large language models for software vulnerability detection: a guide for researchers on models, methods, techniques, datasets, and metrics

International Journal of Information Security, Journal Year: 2025, Volume and Issue: 24(2)

Published: Feb. 14, 2025

Language: Английский

---

A RAG-Based Question-Answering Solution for Cyber-Attack Investigation and Attribution

Lecture notes in computer science, Journal Year: 2025, Volume and Issue: unknown, P. 238 - 256

Published: Jan. 1, 2025

Language: Английский

---

A comparative analysis on using GPT and BERT for automated vulnerability scoring

Intelligent Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 200515 - 200515

Published: April 1, 2025

Language: Английский

---

Enhancing Cybersecurity Situation Awareness Through Knowledge Graphs: An Integrated Survey Including Threats, Vulnerabilities, and Assets

Published: Jan. 1, 2025

Language: Английский

---

Empowering Cyber Defense

Advances in computational intelligence and robotics book series, Journal Year: 2025, Volume and Issue: unknown, P. 429 - 444

Published: May 1, 2025

Academic institutions explore generative artificial intelligence for varied educational and research goals, yet they often operate with limited budgets specialized personnel shortages. Researchers have revealed that models capable of producing sophisticated outputs can be exploited malicious objectives, magnifying cyber risks in settings where funding constraints impede robust defenses. This chapter investigated the impact resource on secure deployment AI, emphasizing emerging exploits stem from data poisoning model inversion. Findings indicated coordinated defense strategies, clear governance frameworks, targeted training mitigate threats without hindering research. These insights suggest collaborative networks practical safeguards are vital sustaining academic innovation. The contributes a framework to help integrate advanced AI capabilities while preserving security.

Language: Английский

---

Classification of Hacker’s Posts Based on Zero-Shot, Few-Shot, and Fine-Tuned LLMs in Environments with Constrained Resources

Future Internet, Journal Year: 2025, Volume and Issue: 17(5), P. 207 - 207

Published: May 5, 2025

This paper investigates, applies, and evaluates state-of-the-art Large Language Models (LLMs) for the classification of posts from a dark web hackers’ forum into four cyber-security categories. The LLMs applied included Mistral-7B-Instruct-v0.2, Gemma-1.1-7B, Llama-3-8B-Instruct, Llama-2-7B, with zero-shot learning, few-shot fine-tuning. categories consisted “Access Control Management”, “Availability Protection Security by Design Mechanisms”, “Software Firmware Flaws”, “not relevant”. were also classified labelled human expert, allowing detailed evaluation accuracy per each LLM customization/learning method. We verified fine-tuning as most effective mechanism to enhance reliability classifications. results include methodology dataset.

Language: Английский

---

Generative AI revolution in cybersecurity: a comprehensive review of threat intelligence and operations

Artificial Intelligence Review, Journal Year: 2025, Volume and Issue: 58(8)

Published: May 7, 2025

Language: Английский

---

Large Language Models for Cyber Resilience: A Comprehensive Review, Challenges, and Future Perspectives

Applied Soft Computing, Journal Year: 2024, Volume and Issue: 170, P. 112663 - 112663

Published: Dec. 25, 2024

Language: Английский

---