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

CrossFit

Benchmarks

Task NameDataset NameSOTA ResultTrend
ClassificationCrossFit cls-23
Accuracy75.4
16
Paraphrase DetectionCrossFit Para (test)
Accuracy66.1
10
Natural Language InferenceCrossFit NLI (test)
Accuracy83.6
10
Text ClassificationCrossFit cls-45 (test)
Accuracy75.2
10
ClassificationCrossFit cls-45
Accuracy77.4
6
Natural Language ProcessingCrossFit (test)
GLUE CoLA Score1
3
Showing 6 of 6 rows