"Academics Develop Testing Benchmark for LLMs in Cyber Threat Intelligence"
"Academics Develop Testing Benchmark for LLMs in Cyber Threat Intelligence"
Rochester Institute of Technology (RIT) researchers created CTIBench, the first benchmark designed for assessing the performance of Large Language Models (LLMs) in Cyber Threat Intelligence (CTI) applications. The researchers emphasized that LLMs could revolutionize CTI by improving security analysts' ability to process and examine massive amounts of unstructured threat and attack data, as well as use more intelligence sources. However, they add that LLMs are vulnerable to hallucinations and text misunderstandings, especially in technical fields.