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

ARC-E

Benchmarks

Task NameDataset NameSOTA ResultTrend
ReasoningMultilingual ARC-E 8 Seen Languages
ARC-E Accuracy85.7
20
Explanation self-consistencyARC-E (test)
Accuracy87.01
4
Showing 2 of 2 rows