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

RoboFine-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Visual Question AnsweringRoboFine-Bench (test)
Overall Accuracy71
6
CaptioningRoboFine-Bench Hard setting
Overall Score83.6
6
CaptioningRoboFine-Bench Easy setting
Overall Score85.2
6
Showing 3 of 3 rows