XDoc: Unified Pre-training for Cross-Format Document Understanding

TL;DR

XDoc is a unified pre-training model that effectively handles multiple document formats with shared parameters and lightweight adaptive layers, achieving comparable or better downstream performance while being cost-efficient.

Contribution

It introduces a single model for various document formats using shared backbone and adaptive layers, improving efficiency and performance over format-specific models.

Findings

XDoc achieves comparable or better performance than format-specific models.

XDoc uses only 36.7% of parameters compared to separate models.

XDoc demonstrates cost-effective deployment for multi-format document understanding.

Abstract

The surge of pre-training has witnessed the rapid development of document understanding recently. Pre-training and fine-tuning framework has been effectively used to tackle texts in various formats, including plain texts, document texts, and web texts. Despite achieving promising performance, existing pre-trained models usually target one specific document format at one time, making it difficult to combine knowledge from multiple document formats. To address this, we propose XDoc, a unified pre-trained model which deals with different document formats in a single model. For parameter efficiency, we share backbone parameters for different formats such as the word embedding layer and the Transformer layers. Meanwhile, we introduce adaptive layers with lightweight parameters to enhance the distinction across different formats. Experimental results have demonstrated that with only 36.7% parameters, XDoc achieves comparable or even better performance on a variety of downstream tasks compared with the individual pre-trained models, which is cost effective for real-world deployment. The code and pre-trained models will be publicly available at \url{https://aka.ms/xdoc}.