Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

BBQ

Benchmarks

Task NameDataset NameSOTA ResultTrend
Bias EvaluationBBQ
Accuracy99.3
99
Question-AnsweringBBQ
Accuracy92.12
36
Question AnsweringBBQ (test)
Accuracy (amb)98.86
20
Bias EvaluationBBQ averaged across gender, nationality, and religion domains
Accuracy (Ambiguous)87.73
16
Question AnsweringBBQ (Bias Benchmark for QA) v1.0 (test)
BBQ SES Score93.1
16
Question AnsweringBBQ
Disambiguation TOP-183.93
12
Question AnsweringBBQ disambiguated questions
Accuracy93
5
Question AnsweringBBQ (ambiguous)
Accuracy95
5
Question Answering Bias EvaluationBBQ
Accuracy (All)79
5
Bias QABBQ Ambig
Accuracy85.04
4
Bias QABBQ Disambig
Accuracy84.85
2
Bias EvaluationBBQ Disambiguated
Bias Score Before90.07
1
Showing 12 of 12 rows