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

Multimodal Reasoning Benchmarks

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal ReasoningMultimodal Reasoning Benchmarks Combined
Overall Score71.86
18
Visual Information Preservation and Explainability EvaluationMultimodal Reasoning Benchmarks (MathVista, WeMath, MMStar, MMMU, R1-Onevision-Bench) (test)
Visual Info Preservation Score3.95
4
Visual Question AnsweringMultimodal Reasoning Benchmarks
FlipVQA Accuracy32.97
2
Showing 3 of 3 rows