vidore/arxivqa_test_subsampled_captioning
Viewer
•
Updated
•
1.44k
•
41
•
1
ViDoRe Captioning (baseline)
The original ViDoRe benchmark was passed to Unstructured to partition each page into chunks. Visual chunks are captioned using Claude Sonnet.
-
-
vidore/docvqa_test_subsampled_captioning
Viewer • Updated • 1.24k • 90 -
vidore/infovqa_test_subsampled_captioning
Viewer • Updated • 2.78k • 63 -
vidore/tabfquad_test_subsampled_captioning
Viewer • Updated • 636 • 41 -
vidore/tatdqa_test_captioning
Viewer • Updated • 8.54k • 58 -
vidore/shiftproject_test_captioning
Viewer • Updated • 2.05k • 39 -
vidore/syntheticDocQA_artificial_intelligence_test_captioning
Viewer • Updated • 1.9k • 65 -
vidore/syntheticDocQA_energy_test_captioning
Viewer • Updated • 2.29k • 50 -
vidore/syntheticDocQA_government_reports_test_captioning
Viewer • Updated • 1.9k • 70 • 1 -
vidore/syntheticDocQA_healthcare_industry_test_captioning
Viewer • Updated • 2.17k • 63