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

Multi-step Visual Reasoning Suite

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-step Visual ReasoningMulti-step Visual Reasoning Suite (DocVQA-RF, TableVQA, VisualProbe, ChartQA-Pro) (train)
Average Tool Calls0.3
7
Showing 1 of 1 rows