ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understanding
This work addresses the challenge of integrating layout knowledge for researchers and practitioners in document AI, offering a novel method that improves performance on specific downstream tasks.
The paper tackles the problem of sub-optimal performance in visually-rich document understanding by proposing ERNIE-Layout, a pre-training solution that enhances layout knowledge, resulting in superior performance and new state-of-the-art results on key information extraction, document image classification, and document question answering datasets.
Recent years have witnessed the rise and success of pre-training techniques in visually-rich document understanding. However, most existing methods lack the systematic mining and utilization of layout-centered knowledge, leading to sub-optimal performances. In this paper, we propose ERNIE-Layout, a novel document pre-training solution with layout knowledge enhancement in the whole workflow, to learn better representations that combine the features from text, layout, and image. Specifically, we first rearrange input sequences in the serialization stage, and then present a correlative pre-training task, reading order prediction, to learn the proper reading order of documents. To improve the layout awareness of the model, we integrate a spatial-aware disentangled attention into the multi-modal transformer and a replaced regions prediction task into the pre-training phase. Experimental results show that ERNIE-Layout achieves superior performance on various downstream tasks, setting new state-of-the-art on key information extraction, document image classification, and document question answering datasets. The code and models are publicly available at http://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-layout.