Cyber defense and threat intel

Part of the benchmark canon: 8 entries, all measure the model. How to read the Paladin column is defined on the canon page.

Benchmark	Organization	Measures	Paladin
CTI-REALM	Microsoft	End-to-end detection-rule generation by agents	Measures the model
ExCyTIn-Bench	Microsoft	Agent-driven cyber threat investigation	Measures the model
CyberSOCEval	Meta with CrowdStrike	Malware analysis and threat-intel reasoning	Measures the model
CTIBench	Academic (Alam et al.)	Threat-intelligence knowledge, attribution, and severity	Measures the model
CyberMetric	TII / Khalifa University	Security knowledge question answering	Measures the model
SecBench	Tencent / HK PolyU	Large multi-domain security question set	Measures the model
SecEval	Academic (Li et al.)	Software, network, and web security questions	Measures the model
SecQA	Academic	Foundational security concepts	Measures the model