List your company

Distributed Misuse Detection on DM-Cyber 20x

84.2Trace-level APNaive Agent
9.63228.99148.3567.709Apr 13, 2026

Evaluation results

MethodLinks
Naive AgentModel=GPT-4o2026.0484.2
MeerkatModel=Qwen2.5-72B2026.0483.7
Naive AgentModel=Qwen2.5-72B2026.0463.2
MeerkatModel=GPT-4o2026.0444.5
BayesianModel=GPT-4o2026.0433.2
BayesianModel=Qwen2.5-72B2026.0424.3
BufferModel=GPT-4o2026.0422.9
MonitorModel=GPT-4o2026.0419.9
BufferModel=Qwen2.5-72B2026.0419.1
MonitorModel=Qwen2.5-72B2026.0412.5