LayoutLMv3-B

Reported on 6 benchmarks across 1 task · 1 paper · 1 SOTA

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

Computer Vision6 results

Document Layout AnalysisonPubLayNet val
Table· 2022-04-18
0.979
best: 0.981 (VGT)
SOTA
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Document Layout AnalysisonPubLayNet val
Figure· 2022-04-18
0.97
best: 0.975 (DETR)
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Document Layout AnalysisonPubLayNet val
List· 2022-04-18
0.955
best: 0.975 (TRDLU)
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Document Layout AnalysisonPubLayNet val
Overall· 2022-04-18
0.951
best: 0.962 (VGT)
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Document Layout AnalysisonPubLayNet val
Text· 2022-04-18
0.945
best: 0.967 (VSR)
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387
Document Layout AnalysisonPubLayNet val
Title· 2022-04-18
0.906
best: 0.939 (VGT)
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking arXiv:2204.08387