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

VStar-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal ReasoningVstar Bench Spatial
Accuracy90.8
19
Multimodal ReasoningVstar Bench Attr
ACC94.8
19
High-resolution Visual UnderstandingVstar Bench
Attribute Score94.8
12
Visual ReasoningVstar Bench Spatial
Accuracy81.6
10
Visual Grounding and ReasoningVStar-Bench
Overall Score84.29
9
Showing 5 of 5 rows