paperSeptember 2026UnreviewedOpen access

Toward Dynamic and Risk-Aware Evaluation of Cybersecurity LLMs: A Survey and the RIRAG Framework

Ravi Prasad, Feroz Ahmed, Shohel Rana, Charan Gudla, Sujan Kumar Reddy Challa, Aditya Garg

Journal of Cybersecurity, Digital Forensics and Jurisprudence

Abstract

The rapid adoption of large language models (LLMs) in cybersecurity has created a growing need for evaluation methods that reflect operational risk rather than isolated language capability. Existing cybersecurity benchmarks assess useful dimensions such as factual knowledge, vulnerability analysis, secure coding, penetration testing, and threat intelligence reasoning, but many remain limited by static datasets, weak diagnostic granularity, limited adversarial testing, and insufficient attention

Cite

@article{prasad2026dynamic,
  title = {{Toward Dynamic and Risk-Aware Evaluation of Cybersecurity LLMs: A Survey and the RIRAG Framework}},
  author = {Ravi Prasad and Feroz Ahmed and Shohel Rana and Charan Gudla and Sujan Kumar Reddy Challa and Aditya Garg},
  year = {2026},
  month = sep,
  journal = {Journal of Cybersecurity, Digital Forensics and Jurisprudence},
  doi = {10.65879/3070-5789.2026.02.05},
  url = {https://www.semanticscholar.org/paper/6d91d3a37230c466cdddb9be0f848f971d28b224}
}