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

CivilComments-WILDS

Benchmarks

Task NameDataset NameSOTA ResultTrend
Group RobustnessCivilComments-WILDS (test)
WG Accuracy67
40
Mean estimationCivilComments-WILDS (test)
Required Labeled Samples1,313
14
Toxicity detectionCivilComments-WILDS (val)
Average Accuracy92.3
5
Showing 3 of 3 rows