Performance Of Neural Machine Translation Research Articles

A Study for Enhancing Low-resource Thai-Myanmar-English Neural Machine Translation

Several methodologies have recently been proposed to enhance the performance of low-resource Neural Machine Translation (NMT). However, these techniques have yet to be explored thoroughly in the low-resource Thai and Myanmar languages. Therefore, we first applied augmentation techniques such as SwitchOut and Ciphertext Based Data Augmentation (CipherDAug) to improve NMT performance in these languages. Second, we enhanced the NMT performance by fine-tuning the pre-trained Multilingual Denoising BART model (mBART), where BART denotes Bidirectional and Auto-Regressive Transformer. We implemented three NMT systems: namely, Transformer+SwitchOut, Multi-Source Transformer+CipherDAug, and fine-tuned mBART in the bidirectional translations of Thai-English-Myanmar language pairs from the ASEAN-MT corpus. Experimental results showed that Multi-Source Transformer+CipherDAug significantly improved Bilingual Evaluation Understudy (BLEU), Character n-gram F-score (ChrF) , and Translation Error Rate (TER) scores over the first baseline Transformer and second baseline Edit-Based Transformer. The model achieved notable BLEU scores: 37.9 (English-to-Thai), 42.7 (Thai-to-English), 28.9 (English-to-Myanmar), 31.2 (Myanmar-to-English), 25.3 (Thai-to-Myanmar), and 25.5 (Myanmar-to-Thai). The fine-tuned mBART model also considerably outperformed the two baselines, except for the Myanmar-to-English pair. SwitchOut improved over the second baseline in all pairs and performed similarly to the first baseline in most cases. Last, we performed detailed analyses verifying that the CipherDAug and mBART models potentially facilitate improving low-resource NMT performance in Thai and Myanmar languages.

Neural machine translation in AVT teaching in China: An in-depth analysis from the readability perspective

As audiovisual translation (AVT) becomes more complex and diverse, the need for advanced machine learning techniques has been increasing sharply, driving the widespread adoption of neural machine translation (NMT) technology in the field. This study contributes to the literature by evaluating the performance of NMT technology in AVT teaching. Based on readability theory, we constructed an evaluation framework with 12 indicators, built comparable corpora consisting of human and post-edited subtitle translations of corporate videos, and used them to examine the performance of four online NMT systems (Google Translate, Baidu Translate, Bing Translator, and Youdao Translate) in AVT teaching. Our statistical analyses and case studies show that Google Translate outperforms the other three platforms in all the readability tests, and it can enhance the readability of post-edited subtitles at five levels (word, syntax, textbase, situation model, genre and rhetorical). The performance of the other three platforms varies across different tests. Concrete examples are provided to substantiate the statistical analyses. Our study adds value to existing research both by examining the application and performance of NMT in AVT teaching and by suggesting potential directions for the refinement of current NMT systems.

A Hybrid Optimization Approach for Neural Machine Translation Using LSTM+RNN with MFO for Under Resource Language (Telugu)

NMT (Neural Machine Translation) is an innovative approach in the field of machine translation, in contrast to SMT (statistical machine translation) and Rule-based techniques which has resulted annotable improvements. This is because NMT is able to overcome many of the shortcomings that are inherent in the traditional approaches. The Development of NMT has grown tremendously in the recent years but NMT performance remain under optimal when applied to low resource language pairs like Telugu, Tamil and Hindi. In this work a proposedmethod fortranslating pairs (Telugu to English) is attempted, an optimal approach which enhancesthe accuracy and execution time period.A hybrid method approach utilizing Long short-term memory (LSTM) and traditional Recurrent Neural Network (RNN) are used for testing and training of the dataset. In the event of long-range dependencies, LSTM will generate more accurate results than a standard RNN would endure and the hybrid technique enhances the performance of LSTM. LSTM is used during the encoding and RNN is used in decoding phases of NMT. Moth Flame Optimization (MFO) is utilized in the proposed system for the purpose of providing the encoder and decoder model with the best ideal points for training the data.

Integrating Reconstructor and Post-Editor into Neural Machine Translation

Neural machine translation (NMT) mainly comprises the encoder and decoder. The encoder is mainly used to extract the feature vector of the source language sentence. The decoder predicts the next token according to the feature vector extracted by the encoder and the information of the current moment. In this process, there is no guarantee that the features extracted by the encoder are indistinguishable from the meaning of the sentences in the source language. There is also no guarantee that the decoder can accurately predict the corresponding character. These issues can lead to over-translation and under-translation issues in the translated results. Previous researchers alleviated this problem by calculating the gap between the reconstructed source-language sentences and the source-language sentences. Inspired by this method, we propose to integrate a reconstructor and a post-editor into NMT during the training. The reconstructor takes the translation of NMT as input to reconstruct the source sentence, and the post-editor takes the translation as input and post-edits it to predict the target sentence. Through the training of the reconstructor and the post-editor, the semantics of the translation are forced to follow the source sentence and the target sentence. Experimental results show that our approach can effectively improve the performance of NMT on multiple translation tasks.

Adding visual attention into encoder-decoder model for multi-modal machine translation

Multi-modal neural machine translation (MNMT) aims to integrate visual and textual information to translate source sentences into target and attracts a lot of attentions. Existing methods contribute a lot for capturing interactions between visual and textual features to improve the performance of neural machine translation (NMT). However, most of them don't consider the multi-modal consistency for MNMT. In fact, the image provides the semantic global consistency between different languages. We believe that adding bilingual-visual agreement into the encoder and decoder simultaneously can obtain bilingual representations and is useful for NMT. In this paper, we propose to simultaneously integrate visual information in encoder and decoder to learn the interactions between visual and textual features, in this paper called the model VMNMT. As the visual information provide global context, the encoder and decoder can learn the bilingual representations. Besides, we introduce a new bilingual-visual agreement decoder to learn to better representations of corresponding image-sentence pairs. In the experiment, the improvement was 2.02 BLEU on the English-German16 dataset and 1.9 BLEU on the English-German17 dataset. The results show that our method can outperform baselines on several widely-used datasets in terms of various metrics.

Unsupervised Parallel Sentences of Machine Translation for Asian Language Pairs

Parallel sentence pairs play a very important role in many natural language processing tasks, especially cross-lingual tasks such as machine translation. So far, many Asian language pairs lack bilingual parallel sentences. As collecting bilingual parallel data is very time-consuming and difficult, it is very important for many low-resource Asian language pairs. While existing methods have shown encouraging results, they rely on bilingual data seriously or have some drawbacks in an unsupervised situation. To address these issues, we propose a new unsupervised similarity calculation and dynamic selection metric to obtain parallel sentence pairs in an unsupervised situation. First, our method maps bilingual word embedding by postdoc adversarial training, which rotates the source space to match the target without parallel data. Then, we introduce a new cross-domain similarity adaption to obtain parallel sentence pairs. Experimental results on real-world datasets show that our model can obtain better accuracy and recall on mining parallel sentence pairs. We also show that the extracted bilingual sentence corpora can significantly improve the performance of neural machine translation.

Efeitos das operações de pré-edição na tradução audiovisual usando o TRADOS: uma análise experimental das traduções de estudantes sauditas

Abstract Neural Machine Translation (NMT) is a revolutionary innovation that has had a significant impact not only on the translation industry but also on translation studies. This technology is creating new opportunities for shaping identities and generating new knowledge and perspectives. This led researchers to undeniably and increasingly conduct studies to improve NMT performance. These studies involve assessing the quality of the output, determining the extent to which humans can be involved in the machine-based process, and proposing solutions to redress the deficiencies. In this regard, the paper focuses on the impact of pre-editing (PE) operations on the translation of audiovisual children’s literary texts (subtitles) from English to Arabic using TRADOS. The work seeks to accomplish a number of pre-editing-related research goals in the context of Arabic NMT systems. Initially, it aims to define and clarify the idea of PE and its function in enhancing Arabic NMT systems’ performance. Second, it looks at PE mechanisms and methods and how they might improve the appropriateness and accuracy of Arabic machine translation. Additionally, the study attempts to investigate the value of PE in enhancing the Arabic translation of audiovisual children’s literature. The study also seeks to list the difficulties and restrictions related to PE and suggest viable strategies to get over them. Finally, the paper offers suggestions for further investigation into NMT systems using Arabic and PE. To achieve these goals, the researchers conducted a human evaluation of NMT for two animated movie subtitles and determined the necessary PE operations. The findings revealed that PE has the potential to improve the quality of literary translations and enhance their comprehensibility and acceptability.

Traduction d’unités polylexicales du portugais en français par MT@EC et eTranslation

Translation of Multi-Word Units from Portuguese into French by MT@EC and eTranslation
 This paper aims to determine the extent to which the shift from statistical machine translation (SMT) to neural machine translation (NMT) improved the performance of European Union machine translation systems between 2015 and 2021 in terms of multi-word unit translation and domain coverage. To do so, we chose to test these systems on machine translation into French of multi-word units expressing quantitative and qualitative progression in Portuguese from Portugal. These units consist of the 2-gram ‘cada vez’ and a comparative adjective or adverb (cada vez COMP), and their word-for-word translation into French is not idiomatic (*chaque fois COMP). The most frequent translation into French is ‘de COMP en COMP’. This implies that these multi-word units must be translated ‘en bloc’, but their identification is not straightforward. On the one hand, COMP is not fixed and may include one (mais / plus, menos / moins, maior / plus grand, menor / plus petit, melhor / meilleur – mieux, pior / pire – plus mal) or several words (mais or menos N, ADJ, ADV). On the other hand, the 2-gram ‘cada vez’ can be part of other multi-word units expressing iteration (de cada vez (que)/(à) chaque fois (que)), or ‘dropper’ ([a certain quantity] de cada vez/ à la fois), This raises the challenge of ambiguity, well known to biotranslators and still often problematic for NMT. Moreover, units expressing quantitative or qualitative progression may raise other translation challenges when they are coordinate (with or without repetition of the 2-gram ‘cada vez’), when they are split (cada vez (…) COMP), or when they combine with verbs or nouns to form extended translation units whose translation into French can result in a more concise solution we refer to as ‘lexicalisation’. We established a biotranslation model based on a manually aligned French-Portuguese parallel literary corpus and online searchable French-Portuguese aligned corpora (translation memories). We selected a sample of occurrences of these multi-word units including several translation challenges. These occurrences were selected from a Portuguese journalistic corpus. They belong therefore to general language, whereas the EU's translation memories cover the domains dealt with by its institutions, which represents an additional challenge, considering the critical importance of domain coverage in the data to NMT performance quality. The selected occurrences were translated into French by the EU SMT system in 2015 (MT@EC) and 2019 (eTranslation Legacy) and by eTranslation (the EU NMT system) in 2019 and 2021. Firstly, MT output was analysed according to two general criteria: ‘non- literality’, that is translation into French without ‘chaque’, and acceptability from a semantic point of view, that is MT output without any false meaning, opposite meaning or nonsense. Then we looked at specific challenges, some of which could lead to original solutions, worthy of a professional human translator, such as lexicalisation, change of grammatical category or ‘recategorisation’ and ‘naturalisation’, that is phraseological or syntactic rearrangement that makes the target text more idiomatic. The results show that MT is improving, especially according to the criterion of non-literality. Original solutions are still rare, but they are diversifying in NMT output. Nevertheless, NMT remains imperfect, not least because of the inherent ambiguity of natural languages and the inevitable gaps in the data on which these systems are based. The results also demonstrate the importance of human intervention in the maintenance of the systems learning automatically, since the quality of SMT system’s output decreases between 2015 and 2019, when all efforts were focused on improving EU NMT system. Finally, results reveal the dangers of using English as a pivot language when translating from one Romance language into another, and the need to train future translators in NMT and post-editing.

Deep Learning-Based English-Chinese Translation Research

Neural machine translation (NMT) has been bringing exciting news in the field of machine translation since its emergence. However, because NMT only employs single neural networks to convert natural languages, it suffers from two drawbacks in terms of reducing translation time: NMT is more sensitive to sentence length than statistical machine translation and the end-to-end implementation process fails to make explicit use of linguistic knowledge to improve translation performance. The network model performance of various deep learning machine translation tasks was constructed and compared in English-Chinese bilingual direction, and the defects of each network were solved by using an attention mechanism. The problems of gradient disappearance and gradient explosion are easy to occur in the recurrent neural network in the long-distance sequence. The short and long-term memory networks cannot reflect the information weight problems in long-distance sequences. In this study, through the comparison of examples, it is concluded that the introduction of an attention mechanism can improve the attention of context information in the process of model generation of the target language sequence, thus translating restore degree and fluency higher. This study proposes a neural machine translation method based on the divide-and-conquer strategy. Based on the idea of divide-and-conquer, this method identifies and extracts the longest noun phrase in a sentence and retains special identifiers or core words to form a sentence frame with the rest of the sentence. This method of translating the longest noun phrase and sentence frame separately by the neural machine translation system, and then recombining the translation, alleviates the poor performance of neural machine translation in long sentences. Experimental results show that the BLEU score of translation obtained by the proposed method has improved by 0.89 compared with the baseline method.

Addressing domain shift in neural machine translation via reinforcement learning

Domain Adaptation (DA) has been a well-known transfer learning algorithm used in Neural Machine Translation (NMT) task. Adding domain-related corpora in training data to train the model boosts the performance of NMT. Many low resource language pairs such as Hindi–Nepali and Spanish–Portuguese, for which sufficient parallel data are not available to train the model, face in-domain data scarcity issue. Non-availability of in-domain training data leads to the use of unrelated corpus in training the model, which degrades the NMT performance. This domain mismatch between training and test data results in a domain shift problem. To tackle the challenge of the domain shift problem in NMT, we propose REINFORCE-based Sentence Selection and Weighting (RSSW) method, which selects pseudo in-domain sentences from out-of-domain data and learns their weights based on Reinforcement Learning. The proposed method leverages the similarity between language pairs by encoding the source and target languages into a common encoding script for language model training. RSSW uses minimum risk training and maximum likelihood estimation as an objective function to train NMT on selected pseudo in-domain sentences. Furthermore, we employ multi-domain and fine-tuning approaches to compare proposed method on Hindi↔Nepali and Hindi↔Marathi language pairs through extensive experimental analysis. From the experimental results, we find that the proposed method outperforms the state-of-the-art and baseline approaches by ∼2 BLEU points in all translation directions of language pairs.

Enhancing low-resource neural machine translation with syntax-graph guided self-attention

Most neural machine translation (NMT) models only rely on parallel sentence pairs, while the performance drops sharply in low-resource cases, as the models fail to mine the linguistry of the corpus. Incorporating prior monolingual knowledge explicitly, such as syntax, has been shown to be effective for NMT, particularly in low-resource scenarios. However, existing approaches have not exploited the full potential of the NMT architectures. In this paper, we present syntax-graph guided self-attention (SGSA): a neural network model that combines the source-side syntactic knowledge with multi-head self-attention. We introduce an additional syntax-aware localness modeling as a bias, which indicates that the syntactically relevant parts need to be paid more attention to. The bias is then incorporated into the original attention distribution to form a revised distribution. Moreover, to maintain the strength of capturing the meaningful semantic representations of source-sentence, we adopt a node random dropping strategy in multi-head self-attention subnetworks. Extensive experiments on several standard small-scale datasets demonstrate that SGSA can significantly improve the performance of Transformer-based NMT, and is also superior to the previous syntax-dependent state-of-the-art.

Improving thai-lao neural machine translation with similarity lexicon

It has been shown that the performance of neural machine translation (NMT) drops starkly in low-resource conditions. Thai-Lao is a typical low-resource language pair of tiny parallel corpus, leading to suboptimal NMT performance on it. However, Thai and Lao have considerable similarities in linguistic morphology and have bilingual lexicon which is relatively easy to obtain. To use this feature, we first build a bilingual similarity lexicon composed of pairs of similar words. Then we propose a novel NMT architecture to leverage the similarity between Thai and Lao. Specifically, besides the prevailing sentence encoder, we introduce an extra similarity lexicon encoder into the conventional encoder-decoder architecture, by which the semantic information carried by the similarity lexicon can be represented. We further provide a simple mechanism in the decoder to balance the information representations delivered from the input sentence and the similarity lexicon. Our approach can fully exploit linguistic similarity carried by the similarity lexicon to improve translation quality. Experimental results demonstrate that our approach achieves significant improvements over the state-of-the-art Transformer baseline system and previous similar works.

Integrating Prior Translation Knowledge Into Neural Machine Translation

Neural machine translation (NMT), which is an encoder-decoder joint neural language model with an attention mechanism, has achieved impressive results on various machine translation tasks in the past several years. However, the language model attribute of NMT tends to produce fluent yet sometimes unfaithful translations, which hinders the improvement of translation capacity. In response to this problem, we propose a simple and efficient method to integrate prior translation knowledge into NMT in a universal manner that is compatible with neural networks. Meanwhile, it enables NMT to consider the crossing language translation knowledge from the source-side of the training pipeline of NMT, thereby making full use of the prior translation knowledge to enhance the performance of NMT. The experimental results on two large-scale benchmark translation tasks demonstrated that our approach achieved a significant improvement over a strong baseline.

TPrune

The invention of Transformer model structure boosts the performance of Neural Machine Translation (NMT) tasks to an unprecedented level. Many previous works have been done to make the Transformer model more execution-friendly on resource-constrained platforms. These researches can be categorized into three key fields: Model Pruning, Transfer Learning, and Efficient Transformer Variants. The family of model pruning methods are popular for their simplicity in practice and promising compression rate and have achieved great success in the field of convolution neural networks (CNNs) for many vision tasks. Nonetheless, previous Transformer pruning works did not perform a thorough model analysis and evaluation on each Transformer component on off-the-shelf mobile devices. In this work, we analyze and prune transformer models at the line-wise granularity and also implement our pruning method on real mobile platforms. We explore the properties of all Transformer components as well as their sparsity features, which are leveraged to guide Transformer model pruning. We name our whole Transformer analysis and pruning pipeline as TPrune. In TPrune, we first propose Block-wise Structured Sparsity Learning (BSSL) to analyze Transformer model property. Then, based on the characters derived from BSSL, we apply Structured Hoyer Square (SHS) to derive the final pruned models. Comparing with the state-of-the-art Transformer pruning methods, TPrune is able to achieve a higher model compression rate with less performance degradation. Experimental results show that our pruned models achieve 1.16×–1.92× speedup on mobile devices with 0%–8% BLEU score degradation compared with the original Transformer model.

Comparing Statistical and Neural Machine Translation Performance on Hindi-To-Tamil and English-To-Tamil

Phrase-based statistical machine translation (PB-SMT) has been the dominant paradigm in machine translation (MT) research for more than two decades. Deep neural MT models have been producing state-of-the-art performance across many translation tasks for four to five years. To put it another way, neural MT (NMT) took the place of PB-SMT a few years back and currently represents the state-of-the-art in MT research. Translation to or from under-resourced languages has been historically seen as a challenging task. Despite producing state-of-the-art results in many translation tasks, NMT still poses many problems such as performing poorly for many low-resource language pairs mainly because of its learning task’s data-demanding nature. MT researchers have been trying to address this problem via various techniques, e.g., exploiting source- and/or target-side monolingual data for training, augmenting bilingual training data, and transfer learning. Despite some success, none of the present-day benchmarks have entirely overcome the problem of translation in low-resource scenarios for many languages. In this work, we investigate the performance of PB-SMT and NMT on two rarely tested under-resourced language pairs, English-To-Tamil and Hindi-To-Tamil, taking a specialised data domain into consideration. This paper demonstrates our findings and presents results showing the rankings of our MT systems produced via a social media-based human evaluation scheme.

Using Sub-character Level Information for Neural Machine Translation of Logographic Languages

Logographic and alphabetic languages (e.g., Chinese vs. English) have different writing systems linguistically. Languages belonging to the same writing system usually exhibit more sharing information, which can be used to facilitate natural language processing tasks such as neural machine translation (NMT). This article takes advantage of the logographic characters in Chinese and Japanese by decomposing them into smaller units, thus more optimally utilizing the information these characters share in the training of NMT systems in both encoding and decoding processes. Experiments show that the proposed method can robustly improve the NMT performance of both “logographic” language pairs (JA–ZH) and “logographic + alphabetic” (JA–EN and ZH–EN) language pairs in both supervised and unsupervised NMT scenarios. Moreover, as the decomposed sequences are usually very long, extra position features for the transformer encoder can help with the modeling of these long sequences. The results also indicate that, theoretically, linguistic features can be manipulated to obtain higher share token rates and further improve the performance of natural language processing systems.

Impact of Filtering Generated Pseudo Bilingual Texts in Low-Resource Neural Machine Translation Enhancement: The Case of Persian-Spanish

Although the Neural Machine Translation (NMT) framework has already been shown effective in large training data scenarios, it is less effective for low-resource conditions. To improve NMT performance in a low-resource setting, we extend the high-quality training data by generating a pseudo bilingual dataset and then filtering out low-quality alignments using a quality estimation based on back-translation. We demonstrate that our approach yields significantly higher BLEU scores than those of a set of baselines.

Explicitation in Neural Machine Translation

This paper is concerned with the following question: to what extent does neural machine translation (NMT) – a relatively new approach to machine translation (MT), which can draw on richer contextual information than previous MT architectures – perform explicitation shifts in translation and how are these shifts realised in linguistic terms? In order to answer this question, the paper attempts to identify instances of explicitation in the machine-translated version of a research report on carbon dioxide capture and storage. The machine-translated text was created using the publicly available generic NMT system DeepL. The human translation of the research report was analysed in a prior research project for instances of explicitation and implicitation (Krüger 2015). After a brief quantitative di scussion of the frequency and distribution of explicitation shifts identified in the DeepL output as compared to the shifts identified in the human translation of the research report, the paper analyses in detail several examples in which DeepL performed explicitation shifts of various kinds. The quantitative and qualitative analyses are intended to yield a tentative picture of the capacity of state-of-the art neural machine translation systems to perform explicitation shifts in translation. As explicitation is understood in this article as an indicator of translational text–context interaction, the explicitation performance of NMT can – to some extent – be taken to be indicative of the “contextual awareness” of this new MT architecture.

Enhanced Neural Machine Translation by Joint Decoding with Word and POS-tagging Sequences

Machine translation has become an irreplaceable application in the use of mobile phones. However, the current mainstream neural machine translation models depend on continuously increasing the amount of parameters to achieve better performance, which is not applicable to the mobile phone. In this paper, we improve the performance of neural machine translation (NMT) with shallow syntax (e.g., POS tag) of target language, which has better accuracy and latency than deep syntax such as dependency parsing. In particular, our models take less parameters and runtime than other complex machine translation models, making mobile applications possible. In detail, we present three RNN-based NMT decoding models (independent decoder, gates shared decoder and fully shared decoder) to jointly predict target word and POS tag sequences. Experiments on Chinese-English and German-English translation tasks show that the fully shared decoder can acquire the best performance, which increases the BLEU score by 1.4 and 2.25 points respectively compared with the attention-based NMT model. In addition, we extend the idea to transformer-based models, and the experimental results also show that the BLEU score is further improved.

Multi-Source Neural Model for Machine Translation of Agglutinative Language

Benefitting from the rapid development of artificial intelligence (AI) and deep learning, the machine translation task based on neural networks has achieved impressive performance in many high-resource language pairs. However, the neural machine translation (NMT) models still struggle in the translation task on agglutinative languages with complex morphology and limited resources. Inspired by the finding that utilizing the source-side linguistic knowledge can further improve the NMT performance, we propose a multi-source neural model that employs two separate encoders to encode the source word sequence and the linguistic feature sequences. Compared with the standard NMT model, we utilize an additional encoder to incorporate the linguistic features of lemma, part-of-speech (POS) tag, and morphological tag by extending the input embedding layer of the encoder. Moreover, we use a serial combination method to integrate the conditional information from the encoders with the outputs of the decoder, which aims to enhance the neural model to learn a high-quality context representation of the source sentence. Experimental results show that our approach is effective for the agglutinative language translation, which achieves the highest improvements of +2.4 BLEU points on Turkish–English translation task and +0.6 BLEU points on Uyghur–Chinese translation task.