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

FiVE-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningFive-Bench Unweighted Mean
Average Score31.06
12
Video EditingFiVE-Bench (test)
Structural Distance10.27
11
Video EditingFiVE-Bench
CLIP-T28.82
8
Object RemovalFiVE-Bench
YN-Acc80
6
Object AdditionFiVE-Bench
YN-Acc100
6
Color AlterationFiVE-Bench
YN Accuracy87
6
Object ReplacementFiVE-Bench
YN-Acc72
6
Showing 7 of 7 rows