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

MMRB2

Benchmarks

Task NameDataset NameSOTA ResultTrend
Preference PredictionMMRB2 out-of-domain
EvalMuse Score70.7
22
Image Generation AssessmentMMRB2 (test)
Accuracy69.2
8
Image Editing EvaluationMMRB2 ImgEdit
Single Score67.2
8
Showing 3 of 3 rows