ResearchBenchmarksAutonomous Cyber Operations on Scenario 1 (Sce1)Follow-0.67Mean RewardCyberOps-Bots-3.8004-2.9877-2.175-1.3623Jan 12, 2026Evaluation ResultsMethodMethodLinksMean RewardReward Coefficient VariationsMean Healthy RatioMean Episode LengthCyberOps-Bots2026.01-0.670.0119197.3MAPPO2026.01-3.170.0577241.2IPPO2026.01-3.440.0598444.1IQL2026.01-3.640.0375923.9QMIX2026.01-3.650.0296020.6VDN2026.01-3.680.0286119.9