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

MultiBLiMP

Benchmarks

Task NameDataset NameSOTA ResultTrend
Linguistic KnowledgeMultiBLiMP (avg)
Accuracy78.8
8
Linguistic ProbingMultiBLiMP (test)
Accuracy (CA)93.3
5
Language UnderstandingMultiBLiMP
Accuracy99
4
Showing 3 of 3 rows