List your company

Malicious Prompt Detection on Babelscape ALERT

99.73AccuracyEnhanced Filtering and Summarization System
8.667632.308855.9579.5912May 2, 2025

Evaluation results

MethodLinks
Enhanced Filtering and Summarization SystemNumber of Prompts=145002025.0599.73
Logistic RegressionNumber of Prompts=145002025.0572.61
Toxic-BERTNumber of Prompts=145002025.0528.81
Hate Speech DetectorNumber of Prompts=145002025.0512.17