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

MaRVL

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multicultural Visual ReasoningMaRVL translated English version (test)
Accuracy73.47
12
Multicultural Visual ReasoningMaRVL
Avg_mul Score62.91
10
Visual ReasoningMaRVL
ID71.66
7
Visual ReasoningMaRVL (test)
Accuracy68.09
7
Showing 4 of 4 rows