BAD
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BAD
80.8F1 Score
11
BAD (test)
37Toxicity Reduction
10
BAD
53.7RSR
9
BAD
66.4RSR
9
BAD (val)
11Toxicity Proportion
7
BAD Against Friend Chat (test)
64.2RSR
7
BAD Against Marv (test)
88.1RSR
7