Unsupervised Neural Machine Translation for Similar and Distant Language Pairs

2021 ◽  
Vol 20 (1) ◽  
pp. 1-17
Author(s):  
Haipeng Sun ◽  
Rui Wang ◽  
Masao Utiyama ◽  
Benjamin Marie ◽  
Kehai Chen ◽  
...  
Keyword(s):  
Machine Translation ◽  
Experimental Results ◽  
Training Data ◽  
Neural Machine Translation ◽  
Efficient Approach ◽  
Latent Representations ◽  
Pseudo Data ◽  
Japanese English ◽  
Better Than ◽  
General Method

Unsupervised neural machine translation (UNMT) has achieved remarkable results for several language pairs, such as French–English and German–English. Most previous studies have focused on modeling UNMT systems; few studies have investigated the effect of UNMT on specific languages. In this article, we first empirically investigate UNMT for four diverse language pairs (French/German/Chinese/Japanese–English). We confirm that the performance of UNMT in translation tasks for similar language pairs (French/German–English) is dramatically better than for distant language pairs (Chinese/Japanese–English). We empirically show that the lack of shared words and different word orderings are the main reasons that lead UNMT to underperform in Chinese/Japanese–English. Based on these findings, we propose several methods, including artificial shared words and pre-ordering, to improve the performance of UNMT for distant language pairs. Moreover, we propose a simple general method to improve translation performance for all these four language pairs. The existing UNMT model can generate a translation of a reasonable quality after a few training epochs owing to a denoising mechanism and shared latent representations. However, learning shared latent representations restricts the performance of translation in both directions, particularly for distant language pairs, while denoising dramatically delays convergence by continuously modifying the training data. To avoid these problems, we propose a simple, yet effective and efficient, approach that (like UNMT) relies solely on monolingual corpora: pseudo-data-based unsupervised neural machine translation. Experimental results for these four language pairs show that our proposed methods significantly outperform UNMT baselines.

Neural machine translation of low-resource languages using SMT phrase pair injection

2020 ◽  
pp. 1-22
Author(s):  
Sukanta Sen ◽  
Mohammed Hasanuzzaman ◽  
Asif Ekbal ◽  
Pushpak Bhattacharyya ◽  
Andy Way
Keyword(s):  
Machine Translation ◽  
Large Scale ◽  
Production Systems ◽  
Statistical Machine Translation ◽  
Training Data ◽  
Original Training ◽  
Neural Machine Translation ◽  
Parallel Corpus ◽  
Low Resource ◽  
Better Than

Abstract Neural machine translation (NMT) has recently shown promising results on publicly available benchmark datasets and is being rapidly adopted in various production systems. However, it requires high-quality large-scale parallel corpus, and it is not always possible to have sufficiently large corpus as it requires time, money, and professionals. Hence, many existing large-scale parallel corpus are limited to the specific languages and domains. In this paper, we propose an effective approach to improve an NMT system in low-resource scenario without using any additional data. Our approach aims at augmenting the original training data by means of parallel phrases extracted from the original training data itself using a statistical machine translation (SMT) system. Our proposed approach is based on the gated recurrent unit (GRU) and transformer networks. We choose the Hindi–English, Hindi–Bengali datasets for Health, Tourism, and Judicial (only for Hindi–English) domains. We train our NMT models for 10 translation directions, each using only 5–23k parallel sentences. Experiments show the improvements in the range of 1.38–15.36 BiLingual Evaluation Understudy points over the baseline systems. Experiments show that transformer models perform better than GRU models in low-resource scenarios. In addition to that, we also find that our proposed method outperforms SMT—which is known to work better than the neural models in low-resource scenarios—for some translation directions. In order to further show the effectiveness of our proposed model, we also employ our approach to another interesting NMT task, for example, old-to-modern English translation, using a tiny parallel corpus of only 2.7K sentences. For this task, we use publicly available old-modern English text which is approximately 1000 years old. Evaluation for this task shows significant improvement over the baseline NMT.

scholarly journals The Development of a Comprehensive Data Set for Systematic Studies of Machine Translation

2021 ◽  
pp. 248-262
Author(s):  
Jörg Tiedemann
Keyword(s):  
Machine Translation ◽  
Training Data ◽  
Sufficient Information ◽  
Data Set ◽  
Neural Machine Translation ◽  
High Resource ◽  
Language Groups ◽  
Comprehensive Data ◽  
Parallel Data ◽  
Initial Results

This paper presents our on-going efforts to develop a comprehensive data set and benchmark for machine translation beyond high-resource languages. The current release includes 500GB of compressed parallel data for almost 3,000 language pairs covering over 500 languages and language variants. We present the structure of the data set and demonstrate its use for systematic studies based on baseline experiments with multilingual neural machine translation between Finno-Ugric languages and other language groups. Our initial results show the capabilities of training effective multilingual translation models with skewed training data but also stress the shortcomings with low-resource settings and the difficulties to obtain sufficient information through straightforward transfer from related languages.

scholarly journals Recurrent Stacking of Layers for Compact Neural Machine Translation Models

2019 ◽  
Vol 33 ◽  
pp. 6292-6299 ◽  
Author(s):  
Raj Dabre ◽  
Atsushi Fujita
Keyword(s):  
Machine Translation ◽  
Single Layer ◽  
Training Data ◽  
Neural Machine Translation ◽  
Parallel Corpora ◽  
Translation Quality ◽  
Sequence Generation ◽  
Sequence Modeling ◽  
Back Translation

In encoder-decoder based sequence-to-sequence modeling, the most common practice is to stack a number of recurrent, convolutional, or feed-forward layers in the encoder and decoder. While the addition of each new layer improves the sequence generation quality, this also leads to a significant increase in the number of parameters. In this paper, we propose to share parameters across all layers thereby leading to a recurrently stacked sequence-to-sequence model. We report on an extensive case study on neural machine translation (NMT) using our proposed method, experimenting with a variety of datasets. We empirically show that the translation quality of a model that recurrently stacks a single-layer 6 times, despite its significantly fewer parameters, approaches that of a model that stacks 6 different layers. We also show how our method can benefit from a prevalent way for improving NMT, i.e., extending training data with pseudo-parallel corpora generated by back-translation. We then analyze the effects of recurrently stacked layers by visualizing the attentions of models that use recurrently stacked layers and models that do not. Finally, we explore the limits of parameter sharing where we share even the parameters between the encoder and decoder in addition to recurrent stacking of layers.

scholarly journals Translating the Molecules: Adapting Neural Machine Translation to Predict IUPAC Names from a Chemical Identifier

2021 ◽  
Author(s):  
Jennifer Handsel ◽  
Brian Matthews ◽  
Nicola Knight ◽  
Simon Coles
Keyword(s):  
Machine Translation ◽  
Inorganic Compounds ◽  
Training Data ◽  
Neural Machine Translation ◽  
Chemical Identifier ◽  
Machine Learning Model ◽  
Decoder Architecture ◽  
Service Training ◽  
The Neural Networks ◽  
Low Coverage

We present a sequence-to-sequence machine learning model for predicting the IUPAC name of a chemical from its standard International Chemical Identifier (InChI). The model uses two stacks of transformers in an encoder-decoder architecture, a setup similar to the neural networks used in state-of-the-art machine translation. Unlike neural machine translation, which usually tokenizes input and output into words or sub-words, our model processes the InChI and predicts the 2 IUPAC name character by character. The model was trained on a dataset of 10 million InChI/IUPAC name pairs freely downloaded from the National Library of Medicine’s online PubChem service. Training took five days on a Tesla K80 GPU, and the model achieved test-set accuracies of 95% (character-level) and 91% (whole name). The model performed particularly well on organics, with the exception of macrocycles. The predictions were less accurate for inorganic compounds, with a character-level accuracy of 71%. This can be explained by inherent limitations in InChI for representing inorganics, as well as low coverage (1.4 %) of the training data.

scholarly journals Unsupervised Neural Machine Translation with SMT as Posterior Regularization

2019 ◽  
Vol 33 ◽  
pp. 241-248 ◽  
Author(s):  
Shuo Ren ◽  
Zhirui Zhang ◽  
Shujie Liu ◽  
Ming Zhou ◽  
Shuai Ma
Keyword(s):  
Machine Translation ◽  
Language Models ◽  
Translation Process ◽  
Weak Supervision ◽  
Neural Machine Translation ◽  
Back Translation ◽  
Negative Effect ◽  
Model Training ◽  
Cross Lingual ◽  
Pseudo Data

Without real bilingual corpus available, unsupervised Neural Machine Translation (NMT) typically requires pseudo parallel data generated with the back-translation method for the model training. However, due to weak supervision, the pseudo data inevitably contain noises and errors that will be accumulated and reinforced in the subsequent training process, leading to bad translation performance. To address this issue, we introduce phrase based Statistic Machine Translation (SMT) models which are robust to noisy data, as posterior regularizations to guide the training of unsupervised NMT models in the iterative back-translation process. Our method starts from SMT models built with pre-trained language models and word-level translation tables inferred from cross-lingual embeddings. Then SMT and NMT models are optimized jointly and boost each other incrementally in a unified EM framework. In this way, (1) the negative effect caused by errors in the iterative back-translation process can be alleviated timely by SMT filtering noises from its phrase tables; meanwhile, (2) NMT can compensate for the deficiency of fluency inherent in SMT. Experiments conducted on en-fr and en-de translation tasks show that our method outperforms the strong baseline and achieves new state-of-the-art unsupervised machine translation performance.

scholarly journals Augmenting Neural Machine Translation through Round-Trip Training Approach

2019 ◽  
Vol 9 (1) ◽  
pp. 268-278 ◽  
Author(s):  
Benyamin Ahmadnia ◽  
Bonnie J. Dorr
Keyword(s):  
Machine Translation ◽  
Training Data ◽  
Training Dataset ◽  
Round Trip ◽  
Neural Machine Translation ◽  
Low Resource ◽  
Translation Quality ◽  
High Resource ◽  
Training Approach ◽  
Language Pair

AbstractThe quality of Neural Machine Translation (NMT), as a data-driven approach, massively depends on quantity, quality and relevance of the training dataset. Such approaches have achieved promising results for bilingually high-resource scenarios but are inadequate for low-resource conditions. Generally, the NMT systems learn from millions of words from bilingual training dataset. However, human labeling process is very costly and time consuming. In this paper, we describe a round-trip training approach to bilingual low-resource NMT that takes advantage of monolingual datasets to address training data bottleneck, thus augmenting translation quality. We conduct detailed experiments on English-Spanish as a high-resource language pair as well as Persian-Spanish as a low-resource language pair. Experimental results show that this competitive approach outperforms the baseline systems and improves translation quality.

scholarly journals Knowledge Graphs Effectiveness in Neural Machine Translation Improvement

2020 ◽  
Vol 21 (3) ◽  
Author(s):  
Benyamin Ahmadnia ◽  
Bonnie J. Dorr ◽  
Parisa Kordjamshidi
Keyword(s):  
Machine Translation ◽  
Semantic Representation ◽  
Language Translation ◽  
Semantic Relations ◽  
Training Data ◽  
Target Language ◽  
Neural Machine Translation ◽  
Source Language ◽  
Knowledge Graphs ◽  
Unknown Words

Neural Machine Translation (NMT) systems require a massive amount of Maintaining semantic relations between words during the translation process yields more accurate target-language output from Neural Machine Translation (NMT). Although difficult to achieve from training data alone, it is possible to leverage Knowledge Graphs (KGs) to retain source-language semantic relations in the corresponding target-language translation. The core idea is to use KG entity relations as embedding constraints to improve the mapping from source to target. This paper describes two embedding constraints, both of which employ Entity Linking (EL)---assigning a unique identity to entities---to associate words in training sentences with those in the KG: (1) a monolingual embedding constraint that supports an enhanced semantic representation of the source words through access to relations between entities in a KG; and (2) a bilingual embedding constraint that forces entity relations in the source-language to be carried over to the corresponding entities in the target-language translation. The method is evaluated for English-Spanish translation exploiting Freebase as a source of knowledge. Our experimental results show that exploiting KG information not only decreases the number of unknown words in the translation but also improves translation quality.

scholarly journals Incorporating Source-Side Phrase Structures into Neural Machine Translation

2019 ◽  
Vol 45 (2) ◽  
pp. 267-292 ◽  
Author(s):  
Akiko Eriguchi ◽  
Kazuma Hashimoto ◽  
Yoshimasa Tsuruoka
Keyword(s):  
Machine Translation ◽  
Syntactic Structure ◽  
Statistical Machine Translation ◽  
Training Data ◽  
Great Success ◽  
Data Set ◽  
Neural Machine Translation ◽  
Proposed Model ◽  
Source Sentence

Neural machine translation (NMT) has shown great success as a new alternative to the traditional Statistical Machine Translation model in multiple languages. Early NMT models are based on sequence-to-sequence learning that encodes a sequence of source words into a vector space and generates another sequence of target words from the vector. In those NMT models, sentences are simply treated as sequences of words without any internal structure. In this article, we focus on the role of the syntactic structure of source sentences and propose a novel end-to-end syntactic NMT model, which we call a tree-to-sequence NMT model, extending a sequence-to-sequence model with the source-side phrase structure. Our proposed model has an attention mechanism that enables the decoder to generate a translated word while softly aligning it with phrases as well as words of the source sentence. We have empirically compared the proposed model with sequence-to-sequence models in various settings on Chinese-to-Japanese and English-to-Japanese translation tasks. Our experimental results suggest that the use of syntactic structure can be beneficial when the training data set is small, but is not as effective as using a bi-directional encoder. As the size of training data set increases, the benefits of using a syntactic tree tends to diminish.

scholarly journals Improving Stylized Neural Machine Translation with Iterative Dual Knowledge Transfer

2021 ◽  
Author(s):  
Xuanxuan Wu ◽  
Jian Liu ◽  
Xinjie Li ◽  
Jinan Xu ◽  
Yufeng Chen ◽  
...  
Keyword(s):  
Knowledge Transfer ◽  
Machine Translation ◽  
Large Scale ◽  
Training Data ◽  
Transfer Model ◽  
Paired Data ◽  
Style Transfer ◽  
Neural Machine Translation ◽  
Translation Model ◽  
Parallel Data

Stylized neural machine translation (NMT) aims to translate sentences of one style into sentences of another style, which is essential for the application of machine translation in a real-world scenario. However, a major challenge in this task is the scarcity of high-quality parallel data which is stylized paired. To address this problem, we propose an iterative dual knowledge transfer framework that utilizes informal training data of machine translation and formality style transfer data to create large-scale stylized paired data, for the training of stylized machine translation model. Specifically, we perform bidirectional knowledge transfer between translation model and text style transfer model iteratively through knowledge distillation. Then, we further propose a data-refinement module to process the noisy synthetic parallel data generated during knowledge transfer. Experiment results demonstrate the effectiveness of our method, achieving an improvement over the existing best model by 5 BLEU points on MTFC dataset. Meanwhile, extensive analyses illustrate our method can also improve the accuracy of formality style transfer.

scholarly journals Automatically Paraphrasing via Sentence Reconstruction and Round-trip Translation

2021 ◽  
Author(s):  
Zilu Guo ◽  
Zhongqiang Huang ◽  
Kenny Q. Zhu ◽  
Guandan Chen ◽  
Kaibo Zhang ◽  
...  
Keyword(s):  
Machine Translation ◽  
Question Answering ◽  
Domain Adaptation ◽  
State Of The Art ◽  
Training Data ◽  
Round Trip ◽  
Previous State ◽  
Supervised Methods ◽  
Paraphrase Generation ◽  
Better Than

Paraphrase generation plays key roles in NLP tasks such as question answering, machine translation, and information retrieval. In this paper, we propose a novel framework for paraphrase generation. It simultaneously decodes the output sentence using a pretrained wordset-to-sequence model and a round-trip translation model. We evaluate this framework on Quora, WikiAnswers, MSCOCO and Twitter, and show its advantage over previous state-of-the-art unsupervised methods and distantly-supervised methods by significant margins on all datasets. For Quora and WikiAnswers, our framework even performs better than some strongly supervised methods with domain adaptation. Further, we show that the generated paraphrases can be used to augment the training data for machine translation to achieve substantial improvements.

Sign in / Sign up

Export Citation Format

Share Document