vidore/arxivqa_test_subsampled_ocr_chunk
Viewer
•
Updated
•
1.44k
•
28
ViDoRe Chunk OCR (baseline)
The ViDoRe benchmark was passed to Unstructured to partition each page into text chunks. Detected figures/tables were captioned with Claude 3-Sonnet.
-
-
vidore/docvqa_test_subsampled_ocr_chunk
Viewer • Updated • 1.24k • 50 -
vidore/infovqa_test_subsampled_ocr_chunk
Viewer • Updated • 2.78k • 29 -
vidore/tabfquad_test_subsampled_ocr_chunk
Viewer • Updated • 636 • 19 -
vidore/tatdqa_test_ocr_chunk
Viewer • Updated • 8.54k • 70 -
vidore/shiftproject_test_ocr_chunk
Viewer • Updated • 2.05k • 34 -
vidore/syntheticDocQA_artificial_intelligence_test_ocr_chunk
Viewer • Updated • 1.9k • 140 -
vidore/syntheticDocQA_energy_test_ocr_chunk
Viewer • Updated • 2.29k • 143 -
vidore/syntheticDocQA_government_reports_test_ocr_chunk
Viewer • Updated • 1.9k • 131 -
vidore/syntheticDocQA_healthcare_industry_test_ocr_chunk
Viewer • Updated • 2.17k • 181