CUTE
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
CUTE 288 samples (test)
99.65Word Accuracy
98
CUTE
96.5Accuracy
92
CUTE (test)
93.5Accuracy
59
CUTE80
99.7Accuracy
59
CUTE Cleaned (test)
99.3Word Accuracy
5
CUTE
51.2Accuracy (CUTE Overall)
3