Towards a Multi-modal, Multi-task Learning based Pre-training Framework for Document Representation Learning
This work addresses the need for better cross-modal and multi-page document processing in AI, though it appears incremental as it builds on existing pre-training and multi-modal techniques.
The paper tackles the problem of learning generic document representations by proposing a multi-task pre-training framework that integrates text, layout, and image modalities for multi-page documents, achieving improved performance on tasks like document classification, information extraction, and retrieval compared to baselines.
Recent approaches in literature have exploited the multi-modal information in documents (text, layout, image) to serve specific downstream document tasks. However, they are limited by their - (i) inability to learn cross-modal representations across text, layout and image dimensions for documents and (ii) inability to process multi-page documents. Pre-training techniques have been shown in Natural Language Processing (NLP) domain to learn generic textual representations from large unlabelled datasets, applicable to various downstream NLP tasks. In this paper, we propose a multi-task learning-based framework that utilizes a combination of self-supervised and supervised pre-training tasks to learn a generic document representation applicable to various downstream document tasks. Specifically, we introduce Document Topic Modelling and Document Shuffle Prediction as novel pre-training tasks to learn rich image representations along with the text and layout representations for documents. We utilize the Longformer network architecture as the backbone to encode the multi-modal information from multi-page documents in an end-to-end fashion. We showcase the applicability of our pre-training framework on a variety of different real-world document tasks such as document classification, document information extraction, and document retrieval. We evaluate our framework on different standard document datasets and conduct exhaustive experiments to compare performance against various ablations of our framework and state-of-the-art baselines.