# Denoising-based UNMT is more robust to word-order divergence than MASS-based UNMT

arXiv: https://arxiv.org/abs/2303.01191 · HTML: https://mlsanity.com/paper/2303.01191/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 25 references that are in the map.
- First posted on arXiv 2 Mar 2023.

## Builds on (most cited first, 13 of 13)

- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [Multilingual Denoising Pre-training for Neural Machine Translation](https://mlsanity.com/paper/2001.08210/) (2020)
- [Improving Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1511.06709/) (2015)
- [MASS: Masked Sequence to Sequence Pre-training for Language Generation](https://mlsanity.com/paper/1905.02450/) (2019)
- [A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddings](https://mlsanity.com/paper/1805.06297/) (2018)
- [Phrase-Based & Neural Unsupervised Machine Translation](https://mlsanity.com/paper/1804.07755/) (2018)
- [Unsupervised Statistical Machine Translation](https://mlsanity.com/paper/1809.01272/) (2018)
- [Word Order Does Matter (And Shuffled Language Models Know It)](https://mlsanity.com/paper/2203.10995/) (2022)
- [Improving the Lexical Ability of Pretrained Language Models for Unsupervised Neural Machine Translation](https://mlsanity.com/paper/2103.10531/) (2021)
- [Addressing word-order Divergence in Multilingual Neural Machine Translation for extremely Low Resource Languages](https://mlsanity.com/paper/1811.00383/) (2018)
- [Crosslingual Embeddings are Essential in UNMT for Distant Languages: An English to IndoAryan Case Study](https://mlsanity.com/paper/2106.04995/) (2021)

## Abstract (arXiv)

We aim to investigate whether UNMT approaches with self-supervised pre-training are robust to word-order divergence between language pairs. We achieve this by comparing two models pre-trained with the same self-supervised pre-training objective. The first model is trained on language pairs with different word-orders, and the second model is trained on the same language pairs with source language re-ordered to match the word-order of the target language. Ideally, UNMT approaches which are robust to word-order divergence should exhibit no visible performance difference between the two configurations. In this paper, we investigate two such self-supervised pre-training based UNMT approaches, namely Masked Sequence-to-Sequence Pre-Training, (MASS) (which does not have shuffling noise) and Denoising AutoEncoder (DAE), (which has shuffling noise). We experiment with five EnglishrightarrowIndic language pairs, i.e., en-hi, en-bn, en-gu, en-kn, and en-ta) where word-order of the source language is SVO (Subject-Verb-Object), and the word-order of the target languages is SOV (Subject-Object-Verb). We observed that for these language pairs, DAE-based UNMT approach consistently outperforms MASS in terms of translation accuracies. Moreover, bridging the word-order gap using reordering improves the translation accuracy of MASS-based UNMT models, while it cannot improve the translation accuracy of DAE-based UNMT models. This observation indicates that DAE-based UNMT is more robust to word-order divergence than MASS-based UNMT. Word-shuffling noise in DAE approach could be the possible reason for the approach being robust to word-order divergence.