MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding

Zilong Wang, Jiuxiang Gu, Chris Tensmeyer, Nikolaos Barmpalios, Ani Nenkova, Tong Sun, Jingbo Shang, Vlad Morariu


Abstract
Document images are a ubiquitous source of data where the text is organized in a complex hierarchical structure ranging from fine granularity (e.g., words), medium granularity (e.g., regions such as paragraphs or figures), to coarse granularity (e.g., the whole page). The spatial hierarchical relationships between content at different levels of granularity are crucial for document image understanding tasks. Existing methods learn features from either word-level or region-level but fail to consider both simultaneously. Word-level models are restricted by the fact that they originate from pure-text language models, which only encode the word-level context. In contrast, region-level models attempt to encode regions corresponding to paragraphs or text blocks into a single embedding, but they perform worse with additional word-level features. To deal with these issues, we propose MGDoc, a new multi-modal multi-granular pre-training framework that encodes page-level, region-level, and word-level information at the same time. MGDoc uses a unified text-visual encoder to obtain multi-modal features across different granularities, which makes it possible to project the multi-granular features into the same hyperspace. To model the region-word correlation, we design a cross-granular attention mechanism and specific pre-training tasks for our model to reinforce the model of learning the hierarchy between regions and words. Experiments demonstrate that our proposed model can learn better features that perform well across granularities and lead to improvements in downstream tasks.
Anthology ID:
2022.emnlp-main.265
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3984–3993
Language:
URL:
https://aclanthology.org/2022.emnlp-main.265
DOI:
Bibkey:
Cite (ACL):
Zilong Wang, Jiuxiang Gu, Chris Tensmeyer, Nikolaos Barmpalios, Ani Nenkova, Tong Sun, Jingbo Shang, and Vlad Morariu. 2022. MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 3984–3993, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding (Wang et al., EMNLP 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2022.emnlp-main.265.pdf