GMN: Generative Multi-modal Network for Practical Document Information Extraction

TL;DR

The paper introduces GMN, a robust multi-modal generative network for document information extraction that handles complex, noisy, and layout-mutable documents without requiring detailed annotations, achieving state-of-the-art results.

Contribution

It presents a novel generative multi-modal approach with spatial encoding and modal-aware masking, capable of processing complex documents without predefined labels or character-level annotations.

Findings

GMN achieves new state-of-the-art performance on public DIE datasets.

It significantly outperforms existing methods in realistic, noisy scenarios.

The approach is robust to OCR errors and complex layouts.

Abstract

Document Information Extraction (DIE) has attracted increasing attention due to its various advanced applications in the real world. Although recent literature has already achieved competitive results, these approaches usually fail when dealing with complex documents with noisy OCR results or mutative layouts. This paper proposes Generative Multi-modal Network (GMN) for real-world scenarios to address these problems, which is a robust multi-modal generation method without predefined label categories. With the carefully designed spatial encoder and modal-aware mask module, GMN can deal with complex documents that are hard to serialized into sequential order. Moreover, GMN tolerates errors in OCR results and requires no character-level annotation, which is vital because fine-grained annotation of numerous documents is laborious and even requires annotators with specialized domain knowledge. Extensive experiments show that GMN achieves new state-of-the-art performance on several public DIE datasets and surpasses other methods by a large margin, especially in realistic scenes.