CyberThreat-Eval: Can Large Language Models Automate Real-World Threat Research?

TL;DR

CyberThreat-Eval is a new benchmark based on real-world threat research workflows that evaluates large language models on practical cybersecurity tasks, highlighting current limitations and guiding future improvements.

Contribution

It introduces a comprehensive, expert-annotated benchmark reflecting real-world CTI workflows and proposes analyst-centric metrics for evaluating LLM performance in cybersecurity.

Findings

LLMs often lack nuanced expertise in complex threat analysis.

Current models struggle to distinguish between correct and incorrect information.

Incorporating human feedback and external databases improves LLM performance.

Abstract

Analyzing Open Source Intelligence (OSINT) from large volumes of data is critical for drafting and publishing comprehensive CTI reports. This process usually follows a three-stage workflow -- triage, deep search and TI drafting. While Large Language Models (LLMs) offer a promising route toward automation, existing benchmarks still have limitations. These benchmarks often consist of tasks that do not reflect real-world analyst workflows. For example, human analysts rarely receive tasks in the form of multiple-choice questions. Also, existing benchmarks often rely on model-centric metrics that emphasize lexical overlap rather than actionable, detailed insights essential for security analysts. Moreover, they typically fail to cover the complete three-stage workflow. To address these issues, we introduce CyberThreat-Eval, which is collected from the daily CTI workflow of a world-leading company. This expert-annotated benchmark assesses LLMs on practical tasks across all three stages as mentioned above. It utilizes analyst-centric metrics that measure factual accuracy, content quality, and operational costs. Our evaluation using this benchmark reveals important insights into the limitations of current LLMs. For example, LLMs often lack the nuanced expertise required to handle complex details and struggle to distinguish between correct and incorrect information. To address these challenges, the CTI workflow incorporates both external ground-truth databases and human expert knowledge. TRA allows human experts to iteratively provide feedback for continuous improvement. The code is available at \href{https://github.com/xschen-beb/CyberThreat-Eval}{\texttt{GitHub}} and \href{https://huggingface.co/datasets/xse/CyberThreat-Eval}{\texttt{HuggingFace}}.