Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

2-2023

Abstract

Alignment between image and text has shown promising im provements on patch-level pre-trained document image mod els. However, investigating more effective or finer-grained alignment techniques during pre-training requires a large amount of computation cost and time. Thus, a question natu rally arises: Could we fine-tune the pre-trained models adap tive to downstream tasks with alignment objectives and achieve comparable or better performance? In this paper, we pro pose a new model architecture with alignment-enriched tuning (dubbed AETNet) upon pre-trained document image models, to adapt downstream tasks with the joint task-specific super vised and alignment-aware contrastive objective. Specifically, weintroduce an extra visual transformer as the alignment-ware image encoder and an extra text transformer as the alignment ware text encoder before multimodal fusion. We consider alignment in the following three aspects: 1) document-level alignment by leveraging the cross-modal and intra-modal con trastive loss; 2) global-local alignment for modeling localized and structural information in document images; and 3) local level alignment for more accurate patch-level information. Ex periments on various downstream tasks show that AETNet can achieve state-of-the-art performance on various downstream tasks. Notably, AETNet consistently outperforms state-of-the art pre-trained models, such as LayoutLMv3 with fine-tuning techniques, on three different downstream tasks. Code is available at https://github.com/MAEHCM/AET.

Discipline

Artificial Intelligence and Robotics | Databases and Information Systems

Publication

AAAI '23: 37th AAAI Conference on Artificial Intelligence, Washington, DC, February 7-14

Volume

37

First Page

2590

Last Page

2598

ISBN

9781577358800

Publisher

AAAI Press

City or Country

Washington

Copyright Owner and License

Authors

Download

Share

COinS