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

m-MMLU

Benchmarks

Task NameDataset NameSOTA ResultTrend
Cross-lingual Alignment Correlationm-MMLU FLORES (test)
Pearson Correlation0.9859
81
Multilingual UnderstandingM-MMLU
Normalized Performance1.235
13
Multiple-choice Question Answeringm-MMLU English
Accuracy77
9
Pearson correlation analysism-MMLU
Pearson Correlation (r)0.9849
6
Showing 4 of 4 rows