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

Image Implication

Benchmarks

Task NameDataset NameSOTA ResultTrend
Open-Style Question AnsweringImage Implication Open-Style 1.0 (test)
Score3.94
25
Multiple-Choice Question AnsweringImage Implication Multiple-Choice 1.0 (test)
Accuracy78
25
True-False Question AnsweringImage Implication True-False 1.0 (test)
Accuracy74
25
Showing 3 of 3 rows