An Experimental Study of Deep Neural Network Models for Vietnamese Multiple-Choice Reading Comprehension

TL;DR

This paper investigates the effectiveness of neural network models, including BERT, for Vietnamese multiple-choice reading comprehension, highlighting the impact of word representations in low-resource language settings.

Contribution

It provides the first extensive experimental analysis of neural models and word embeddings for Vietnamese MRC, demonstrating BERT's competitive performance.

Findings

BERT achieved 61.28% accuracy on the ViMMRC test set.

Word representation significantly affects model performance in Vietnamese MRC.

Experiments with the Co-match model on various embeddings reveal varying impacts on accuracy.

Abstract

Machine reading comprehension (MRC) is a challenging task in natural language processing that makes computers understanding natural language texts and answer questions based on those texts. There are many techniques for solving this problems, and word representation is a very important technique that impact most to the accuracy of machine reading comprehension problem in the popular languages like English and Chinese. However, few studies on MRC have been conducted in low-resource languages such as Vietnamese. In this paper, we conduct several experiments on neural network-based model to understand the impact of word representation to the Vietnamese multiple-choice machine reading comprehension. Our experiments include using the Co-match model on six different Vietnamese word embeddings and the BERT model for multiple-choice reading comprehension. On the ViMMRC corpus, the accuracy of BERT model is 61.28% on test set.