Advancing Question Answering on Handwritten Documents: A State-of-the-Art Recognition-Based Model for HW-SQuAD

TL;DR

This paper introduces a recognition-based model for question answering on handwritten documents, utilizing transformer retrieval and ensemble methods to significantly improve accuracy on HW-SQuAD and BenthamQA datasets.

Contribution

It presents a novel recognition-based approach with transformer retrieval and ensemble techniques that outperforms previous state-of-the-art models in handwritten document question answering.

Findings

Achieved 82.02% Exact Match on HW-SQuAD

Boosted document retrieval accuracy to 95.30%

Surpassed previous best recognition-based methods by significant margins

Abstract

Question-answering handwritten documents is a challenging task with numerous real-world applications. This paper proposes a novel recognition-based approach that improves upon the previous state-of-the-art on the HW-SQuAD and BenthamQA datasets. Our model incorporates transformer-based document retrieval and ensemble methods at the model level, achieving an Exact Match score of 82.02% and 69% in HW-SQuAD and BenthamQA datasets, respectively, surpassing the previous best recognition-based approach by 10.89% and 3%. We also enhance the document retrieval component, boosting the top-5 retrieval accuracy from 90% to 95.30%. Our results demonstrate the significance of our proposed approach in advancing question answering on handwritten documents. The code and trained models will be publicly available to facilitate future research in this critical area of natural language.