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

WSC

Benchmarks

Task NameDataset NameSOTA ResultTrend
Coreference ResolutionWSC
Accuracy98.5
116
ClassificationWSC
Accuracy80
65
Coreference ResolutionWSC
Accuracy@185.2
33
LanguageWSC
Score86.51
30
Coreference ResolutionWSC
Loss0.02
20
Coreference ResolutionWSC (test)
Accuracy82.7
19
Pronoun DisambiguationWSC (test)
Accuracy (Single)78.8
14
Coreference ResolutionWSC
Accuracy65.4
13
Commonsense ReasoningWSC
Accuracy80.6
12
Coreference ResolutionWSC
Accuracy79.2
11
ReasoningWSC Ambiguity-Augmented (200 samples)
Accuracy@185.2
11
Coreference ResolutionWSC ambiguity-augmented
Accuracy82.6
11
Coreference ResolutionWSC
Exact Match (EM)116
8
Winograd Schema ChallengeWSC
Accuracy43.3
8
Coreference ResolutionWSC SuperGLUE (test)
Accuracy (Test)65.65
8
Coreference ResolutionWSC standard (test)
Accuracy56.7
8
Sleep stagingWSC
AUC98.1
7
Coreference ResolutionWSC
F1 Score58.36
7
Coreference ResolutionWSC
Accuracy (0-shot)75.7
6
Coreference ResolutionWSC (dev)
Accuracy85.6
6
Coreference ResolutionWSC273
Accuracy82.8
5
Showing 21 of 21 rows