LayoutMask: Enhance Text-Layout Interaction in Multi-modal Pre-training for Document Understanding
Visually-rich Document Understanding (VrDU) has attracted much research attention over the past years. Pre-trained models on a large number of document images with transformer-based backbones have led to significant performance gains in this field. The major challenge is how to fusion the different modalities (text, layout, and image) of the documents in a unified model with different pre-training tasks. This paper focuses on improving text-layout interactions and proposes a novel multi-modal pre-training model, LayoutMask. LayoutMask uses local 1D position, instead of global 1D position, as layout input and has two pre-training objectives: (1) Masked Language Modeling: predicting masked tokens with two novel masking strategies; (2) Masked Position Modeling: predicting masked 2D positions to improve layout representation learning. LayoutMask can enhance the interactions between text and layout modalities in a unified model and produce adaptive and robust multi-modal representations for downstream tasks. Experimental results show that our proposed method can achieve state-of-the-art results on a wide variety of VrDU problems, including form understanding, receipt understanding, and document image classification.
PDF AbstractTask | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
---|---|---|---|---|---|---|
Key Information Extraction | CORD | LayoutMask (base) | F1 | 96.99 | # 4 | |
Key Information Extraction | CORD | LayoutMask (large) | F1 | 97.19 | # 3 | |
Named Entity Recognition (NER) | CORD-r | LayoutMask | F1 | 81.84 | # 4 | |
Semantic entity labeling | FUNSD | LayoutMask (large) | F1 | 93.20 | # 1 | |
Semantic entity labeling | FUNSD | LayoutMask (base) | F1 | 92.91 | # 3 | |
Named Entity Recognition (NER) | FUNSD-r | LayoutMask | F1 | 77.10 | # 4 |