List your company

Distributed Misuse Detection on DM-Bio 20x

59.3Trace-level APMeerkat
2.82817.48932.1546.811Apr 13, 2026

Evaluation results

MethodLinks
MeerkatModel=Qwen2.5-72B2026.0459.3
Naive AgentModel=Qwen2.5-72B2026.0457.2
BayesianModel=GPT-4o2026.0424.6
BayesianModel=Qwen2.5-72B2026.0417.6
BufferModel=Qwen2.5-72B2026.0414.2
MeerkatModel=GPT-4o2026.0410
BufferModel=GPT-4o2026.049.5
MonitorModel=Qwen2.5-72B2026.045.4
Naive AgentModel=GPT-4o2026.045