0

Visually Guided Generative Text-Layout Pre-training for Document Intelligence

ViTLP, a visually guided generative text-layout pre-training model, achieves competitive performance in VDU tasks by optimizing hierarchical language and layout modeling objectives and using a multi-segment pre-training scheme.

Year
2024
Venue
arXiv 2024
Authors
7
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2403.16516v2ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

Prior study shows that pre-training techniques can boost the performance of visual document understanding (VDU), which typically requires models to gain abilities to perceive and reason both document texts and layouts (e.g., locations of texts and table-cells). To this end, we propose visually guided generative text-layout pre-training, named ViTLP. Given a document image, the model optimizes hierarchical language and layout modeling objectives to generate the interleaved text and layout sequence. In addition, to address the limitation of processing long documents by Transformers, we introduce a straightforward yet effective multi-segment generative pre-training scheme, facilitating ViTLP to process word-intensive documents of any length. ViTLP can function as a native OCR model to localize and recognize texts of document images. Besides, ViTLP can be effectively applied to various downstream VDU tasks. Extensive experiments show that ViTLP achieves competitive performance over existing baselines on benchmark VDU tasks, including information extraction, document classification, and document question answering.

Authors

7