Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agent Traces

Benchmarks

Task NameDataset NameSOTA ResultTrend
Attack DetectionAgent Traces IID
ROC AUC0.93
3
Attack DetectionAgent Traces (Held-out)
ROC-AUC (Social engineering)0.89
3
Showing 2 of 2 rows