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

WaNLI

Benchmarks

Task NameDataset NameSOTA ResultTrend
Natural Language InferenceWANLI
Accuracy (WANLI)45.4
42
Natural Language InferenceWANLI (test)
Accuracy96.84
21
Natural Language InferenceWaNLI (OOD)
Accuracy59.86
4
Showing 3 of 3 rows