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

Safety Evaluation Scenarios

Benchmarks

Task NameDataset NameSOTA ResultTrend
Safety ClassificationSafety Evaluation Scenarios Government Decision
Safety Accuracy (Safe)100
2
Safety ClassificationSafety Evaluation Scenarios Health Consultation
Safe Rate97
2
Safety ClassificationSafety Evaluation Scenarios Financial Advice
Safe Accuracy99.6
2
Safety ClassificationSafety Evaluation Scenarios Legal Opinion
Safe Rate95
2
Safety ClassificationSafety Evaluation Scenarios Privacy Violence
Safety Rate98.6
2
Safety ClassificationSafety Evaluation Scenarios Political Lobbying
Safe Accuracy100
2
Safety ClassificationSafety Evaluation Scenarios Pornography
Safe Rate98.9
2
Safety ClassificationSafety Evaluation Scenarios Fraud
Safe Rate99.9
2
Safety ClassificationSafety Evaluation Scenarios Economic Harm
Safe Rate100
2
Safety ClassificationSafety Evaluation Scenarios Physical Harm
Safe Rate100
2
Safety ClassificationSafety Evaluation Scenarios Malware
Safety Accuracy98
2
Safety ClassificationSafety Evaluation Scenarios Hate Speech
Safe Classification Rate99.9
2
Safety ClassificationSafety Evaluation Scenarios Illegal Activity
Safety Rate99.8
2
Showing 13 of 13 rows