# Handling Syntactic Divergence in Low-resource Machine Translation

arXiv: https://arxiv.org/abs/1909.00040 · HTML: https://mlsanity.com/paper/1909.00040/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Aug 2019.

## Built on by (most cited first, 8 of 8)

- [Data Augmentation for Sign Language Gloss Translation](https://mlsanity.com/paper/2105.07476/) (2021)
- [A Survey on Low-Resource Neural Machine Translation](https://mlsanity.com/paper/2107.04239/) (2021)
- [Universal Multimodal Representation for Language Understanding](https://mlsanity.com/paper/2301.03344/) (2023)
- [IndT5: A Text-to-Text Transformer for 10 Indigenous Languages](https://mlsanity.com/paper/2104.07483/) (2021)
- [To Diverge or Not to Diverge: A Morphosyntactic Perspective on Machine Translation vs Human Translation](https://mlsanity.com/paper/2401.01419/) (2024)
- [Bridging Cross-Lingual Gaps During Leveraging the Multilingual Sequence-to-Sequence Pretraining for Text Generation and Understanding](https://mlsanity.com/paper/2204.07834/) (2022)
- [Enhancing the Transformer Decoder with Transition-based Syntax](https://mlsanity.com/paper/2101.12640/) (2021)
- [Linguistically-driven Multi-task Pre-training for Low-resource Neural Machine Translation](https://mlsanity.com/paper/2201.08070/) (2022)

## Abstract (arXiv)

Despite impressive empirical successes of neural machine translation (NMT) on standard benchmarks, limited parallel data impedes the application of NMT models to many language pairs. Data augmentation methods such as back-translation make it possible to use monolingual data to help alleviate these issues, but back-translation itself fails in extreme low-resource scenarios, especially for syntactically divergent languages. In this paper, we propose a simple yet effective solution, whereby target-language sentences are re-ordered to match the order of the source and used as an additional source of training-time supervision. Experiments with simulated low-resource Japanese-to-English, and real low-resource Uyghur-to-English scenarios find significant improvements over other semi-supervised alternatives.