A neural joint model for Vietnamese word segmentation, POS tagging and dependency parsing

TL;DR

This paper introduces a multi-task neural model that jointly performs Vietnamese word segmentation, POS tagging, and dependency parsing, achieving state-of-the-art results on benchmark datasets.

Contribution

It is the first multi-task learning approach for these three Vietnamese NLP tasks integrated into a single neural model.

Findings

Achieved state-of-the-art performance on Vietnamese benchmarks.

Demonstrated the effectiveness of joint learning for multiple NLP tasks.

Outperformed previous separate models in accuracy.

Abstract

We propose the first multi-task learning model for joint Vietnamese word segmentation, part-of-speech (POS) tagging and dependency parsing. In particular, our model extends the BIST graph-based dependency parser (Kiperwasser and Goldberg, 2016) with BiLSTM-CRF-based neural layers (Huang et al., 2015) for word segmentation and POS tagging. On Vietnamese benchmark datasets, experimental results show that our joint model obtains state-of-the-art or competitive performances.