Syntax-Aware Complex-Valued Neural Machine Translation

TL;DR

This paper introduces a syntax-aware complex-valued neural machine translation model that jointly learns syntax and word attention, improving translation quality especially for language pairs with different syntax.

Contribution

It presents a novel, architecture-independent method to incorporate syntax into complex-valued NMT models using joint attention mechanisms.

Findings

Significant BLEU score improvements on two datasets.

Greater gains for language pairs with syntactic differences.

Model is compatible with existing Seq2Seq frameworks.

Abstract

Syntax has been proven to be remarkably effective in neural machine translation (NMT). Previous models obtained syntax information from syntactic parsing tools and integrated it into NMT models to improve translation performance. In this work, we propose a method to incorporate syntax information into a complex-valued Encoder-Decoder architecture. The proposed model jointly learns word-level and syntax-level attention scores from the source side to the target side using an attention mechanism. Importantly, it is not dependent on specific network architectures and can be directly integrated into any existing sequence-to-sequence (Seq2Seq) framework. The experimental results demonstrate that the proposed method can bring significant improvements in BLEU scores on two datasets. In particular, the proposed method achieves a greater improvement in BLEU scores in translation tasks involving language pairs with significant syntactic differences.