Survey on Question Answering over Visually Rich Documents: Methods, Challenges, and Trends

TL;DR

This survey reviews current methods, challenges, and trends in question answering over visually-rich documents, highlighting the need for standardized processing pipelines and future research directions.

Contribution

It offers a comprehensive overview of state-of-the-art approaches, analyzing their strengths and limitations, and identifies key challenges and promising research directions in the field.

Findings

Highlights the lack of consensus on processing pipelines

Identifies main challenges in visually-rich document understanding

Proposes promising future research directions

Abstract

The field of visually-rich document understanding, which involves interacting with visually-rich documents (whether scanned or born-digital), is rapidly evolving and still lacks consensus on several key aspects of the processing pipeline. In this work, we provide a comprehensive overview of state-of-the-art approaches, emphasizing their strengths and limitations, pointing out the main challenges in the field, and proposing promising research directions.