Donut

Reported on 3 benchmarks across 2 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing3 results

Visual Question Answering (VQA)onDocVQA test
ANLS· 2021-11-30
0.675
best: 0.9436 (Human)
OCR-free Document Understanding Transformer arXiv:2111.15664
Key Information ExtractiononRFUND-EN
key-value pair F1· 2021-11-30
24.54
best: 79.27 (PEneo (LayoutLMv3_base))
OCR-free Document Understanding Transformer arXiv:2111.15664
Key Information ExtractiononSIBR
key-value pair F1· 2021-11-30
17.26
best: 82.52 (PEneo (LayoutLMv3_base_chinese))
OCR-free Document Understanding Transformer arXiv:2111.15664