Syntax-Aware Complex-Valued Neural Machine Translation
This work addresses translation quality for language pairs with significant syntactic differences, but it is incremental as it builds on existing syntax-aware NMT methods.
The authors tackled the problem of improving neural machine translation by incorporating syntax information into a complex-valued encoder-decoder architecture, resulting in significant BLEU score improvements on two datasets, especially for language pairs with large syntactic differences.
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.