Unsupervised Translation of German--Lower Sorbian: Exploring Training and Novel Transfer Methods on a Low-Resource Language

• URL: http://arxiv.org/abs/2109.12012v1
• Date: Fri, 24 Sep 2021 15:11:22 GMT
• Title: Unsupervised Translation of German--Lower Sorbian: Exploring Training and Novel Transfer Methods on a Low-Resource Language
• Authors: Lukas Edman, Ahmet \"Ust\"un, Antonio Toral, Gertjan van Noord
• Abstract summary: This paper describes the methods behind the systems submitted by the University of Groningen for the WMT 2021 Unsupervised Machine Translation task. Our system uses a transformer encoder-decoder architecture in which we make three changes to the standard training procedure. We introduce a novel method for initializing the vocabulary of an unseen language, achieving improvements of 3.2 BLEU for DE$rightarrow$DSB and 4.0 BLEU for DSB$rightarrow$DE.
• Score: 2.4870937127982344
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper describes the methods behind the systems submitted by the University of Groningen for the WMT 2021 Unsupervised Machine Translation task for German--Lower Sorbian (DE--DSB): a high-resource language to a low-resource one. Our system uses a transformer encoder-decoder architecture in which we make three changes to the standard training procedure. First, our training focuses on two languages at a time, contrasting with a wealth of research on multilingual systems. Second, we introduce a novel method for initializing the vocabulary of an unseen language, achieving improvements of 3.2 BLEU for DE$\rightarrow$DSB and 4.0 BLEU for DSB$\rightarrow$DE. Lastly, we experiment with the order in which offline and online back-translation are used to train an unsupervised system, finding that using online back-translation first works better for DE$\rightarrow$DSB by 2.76 BLEU. Our submissions ranked first (tied with another team) for DSB$\rightarrow$DE and third for DE$\rightarrow$DSB.

Related papers

• Breaking the Script Barrier in Multilingual Pre-Trained Language Models with Transliteration-Based Post-Training Alignment [50.27950279695363]
  The transfer performance is often hindered when a low-resource target language is written in a different script than the high-resource source language. Inspired by recent work that uses transliteration to address this problem, our paper proposes a transliteration-based post-pretraining alignment (PPA) method.
  arXiv  Detail & Related papers  (2024-06-28T08:59:24Z)
• Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translation [53.22775597051498]
  We present a continual pre-training framework on mBART to effectively adapt it to unseen languages. Results show that our method can consistently improve the fine-tuning performance upon the mBART baseline. Our approach also boosts the performance on translation pairs where both languages are seen in the original mBART's pre-training.
  arXiv  Detail & Related papers  (2021-05-09T14:49:07Z)
• Unsupervised Transfer Learning in Multilingual Neural Machine Translation with Cross-Lingual Word Embeddings [72.69253034282035]
  We exploit a language independent multilingual sentence representation to easily generalize to a new language. Blindly decoding from Portuguese using a basesystem containing several Romance languages we achieve scores of 36.4 BLEU for Portuguese-English and 12.8 BLEU for Russian-English. We explore a more practical adaptation approach through non-iterative backtranslation, exploiting our model's ability to produce high quality translations.
  arXiv  Detail & Related papers  (2021-03-11T14:22:08Z)
• The LMU Munich System for the WMT 2020 Unsupervised Machine Translation Shared Task [125.06737861979299]
  This paper describes the submission of LMU Munich to the WMT 2020 unsupervised shared task, in two language directions. Our core unsupervised neural machine translation (UNMT) system follows the strategy of Chronopoulou et al. We ensemble our best-performing systems and reach a BLEU score of 32.4 on German->Upper Sorbian and 35.2 on Upper Sorbian->German.
  arXiv  Detail & Related papers  (2020-10-25T19:04:03Z)
• SJTU-NICT's Supervised and Unsupervised Neural Machine Translation Systems for the WMT20 News Translation Task [111.91077204077817]
  We participated in four translation directions of three language pairs: English-Chinese, English-Polish, and German-Upper Sorbian. Based on different conditions of language pairs, we have experimented with diverse neural machine translation (NMT) techniques. In our submissions, the primary systems won the first place on English to Chinese, Polish to English, and German to Upper Sorbian translation directions.
  arXiv  Detail & Related papers  (2020-10-11T00:40:05Z)
• Reusing a Pretrained Language Model on Languages with Limited Corpora for Unsupervised NMT [129.99918589405675]
  We present an effective approach that reuses an LM that is pretrained only on the high-resource language. The monolingual LM is fine-tuned on both languages and is then used to initialize a UNMT model. Our approach, RE-LM, outperforms a competitive cross-lingual pretraining model (XLM) in English-Macedonian (En-Mk) and English-Albanian (En-Sq)
  arXiv  Detail & Related papers  (2020-09-16T11:37:10Z)
• Cross-model Back-translated Distillation for Unsupervised Machine Translation [21.79719281036467]
  We introduce a novel component to the standard UMT framework called Cross-model Back-translated Distillation (CBD) CBD achieves the state of the art in the WMT'14 English-French, WMT'16 English-German and English-Romanian bilingual unsupervised translation tasks. It also yields 1.5-3.3 BLEU improvements in IWSLT English-French and English-German tasks.
  arXiv  Detail & Related papers  (2020-06-03T10:57:21Z)
• Neural Machine Translation for Low-Resourced Indian Languages [4.726777092009554]
  Machine translation is an effective approach to convert text to a different language without any human involvement. In this paper, we have applied NMT on two of the most morphological rich Indian languages, i.e. English-Tamil and English-Malayalam. We proposed a novel NMT model using Multihead self-attention along with pre-trained Byte-Pair-Encoded (BPE) and MultiBPE embeddings to develop an efficient translation system.
  arXiv  Detail & Related papers  (2020-04-19T17:29:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.