Dr. DocBench evaluates expert-level document parsing on difficult multilingual pages with layout, reading-order, and domain-specific annotations.
unassessed
| Category | multimodal |
|---|---|
| Subcategory | document parsing |
| Metric | parsing quality |
| Direction | higher_is_better |
| Unit | score |
| Dataset size | 4514 |
| Publisher | Dr. DocBench authors |
Dr. DocBench tests vision-language models and document parsers on challenging pages from long multilingual books. It spans 52 BISAC subject domains and targets chemical formulae, music notation, complex tables, cross-page layouts, and other structures where modern parsers fail.
Document-page parsing with page- and block-level structural annotations.
No model card in ModelSpec reports this benchmark yet.