LayoutLM: Pre-training of Text and Layout for Document Image Understanding

Open paper
Benchmark Model Rank Results
relation-extraction-on-funsdLayoutLM#8F1: 42.83
document-image-classification-on-rvl-cdipPre-trained LayoutLM#15Accuracy: 94.42%Parameters: 160M
semantic-entity-labeling-on-funsdLayoutLM Base#17F1: 79.27

Rank counts only results with a code link.