BoLD
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BOLD
1.9Toxicity (Max)
28
BOLD
0Toxic Rate
26
BOLD 23679 prompts (test)
0.02Avg Toxicity (Max)
18
BOLD
50.3BOLD Score
17
BoLD
26.66mAP
8
BOLD
0.016Toxicity Score (All)
5
BOLD
0.066Gender
4
BOLD (test)
0.05Bias Score (Gender)
4
BOLD
1.037Bias Score
4
BoLD official (test)
0.16mR2
3