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

BoLD

Benchmarks

Task NameDataset NameSOTA ResultTrend
DetoxificationBOLD
Toxicity (Max)1.9
28
Toxicity EvaluationBOLD
Toxic Rate0
26
Toxicity EvaluationBOLD 23679 prompts (test)
Avg Toxicity (Max)0.02
18
Bias and Sentiment EvaluationBOLD
BOLD Score50.3
17
Emotion RecognitionBoLD
mAP26.66
8
Language Generation Bias EvaluationBOLD
Toxicity Score (All)0.016
5
Regard EvaluationBOLD
Gender0.0663
4
Bias EvaluationBOLD (test)
Bias Score (Gender)0.0502
4
Bias EvaluationBOLD
Bias Score1.037
4
Emotion RecognitionBoLD official (test)
mR20.1597
3
Showing 10 of 10 rows