LayoutLMv3

Reported on 5 benchmarks across 3 tasks · 1 paper · 5 SOTA

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

Natural Language Processing5 results

Named Entity Recognition (NER)onFUNSD-r
F1· 2022-04-18
78.77
best: 80.4 (TPP (LayoutLMv3))
SOTA
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Named Entity Recognition (NER)onCORD-r
F1· 2022-04-18
82.72
best: 91.85 (TPP (LayoutLMv3))
SOTA
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Document AIonEPHOIE
Average F1· 2022-04-18
99.21
SOTA
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Key Information ExtractiononEPHOIE
Average F1· 2022-04-18
99.21
SOTA
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Key Information ExtractiononRFUND-EN
key-value pair F1· 2022-04-18
57.66
best: 79.27 (PEneo (LayoutLMv3_base))
SOTA
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387