Revise: A Framework for Revising OCRed text in Practical Information Systems with Data Contamination Strategy

TL;DR

Revise is a framework that systematically corrects OCR errors at multiple levels using synthetic data, improving document understanding and management in Document AI applications.

Contribution

It introduces a hierarchical taxonomy of OCR errors and a synthetic data generation strategy to train effective correction models for structured document understanding.

Findings

Revise significantly improves OCR correction accuracy.

Enhanced OCR correction leads to better document retrieval and question answering performance.

The framework effectively manages structural errors in OCR outputs.

Abstract

Recent advances in Large Language Models (LLMs) have significantly improved the field of Document AI, demonstrating remarkable performance on document understanding tasks such as question answering. However, existing approaches primarily focus on solving specific tasks, lacking the capability to structurally organize and manage document information. To address this limitation, we propose Revise, a framework that systematically corrects errors introduced by OCR at the character, word, and structural levels. Specifically, Revise employs a comprehensive hierarchical taxonomy of common OCR errors and a synthetic data generation strategy that realistically simulates such errors to train an effective correction model. Experimental results demonstrate that Revise effectively corrects OCR outputs, enabling more structured representation and systematic management of document contents. Consequently, our method significantly enhances downstream performance in document retrieval and question answering tasks, highlighting the potential to overcome the structural management limitations of existing Document AI frameworks.