What do Neural Machine Translation Models Learn about Morphology?

Most dominant neural machine translation (NMT) models are restricted to make predictions only according to the local context of preceding words in a left-to-right manner. Although many previous studies try to incorporate global information into NMT models, there still exist limitations on how to effectively exploit bidirectional global context. In this paper, we propose a Confidence Based Bidirectional Global Context Aware (CBBGCA) training framework for NMT, where the NMT model is jointly trained with an auxiliary conditional masked language model (CMLM). The training consists of two stages: (1) multi-task joint training; (2) confidence based knowledge distillation. At the first stage, by sharing encoder parameters, the NMT model is additionally supervised by the signal from the CMLM decoder that contains bidirectional global contexts. Moreover, at the second stage, using the CMLM as teacher, we further pertinently incorporate bidirectional global context to the NMT model on its unconfidently-predicted target words via knowledge distillation. Experimental results show that our proposed CBBGCA training framework significantly improves the NMT model by +1.02, +1.30 and +0.57 BLEU scores on three large-scale translation datasets, namely WMT'14 English-to-German, WMT'19 Chinese-to-English and WMT'14 English-to-French, respectively.

In this paper we describe our team's (NICT-5) Neural Machine Translation (NMT) models whose translations were submitted to shared tasks of the 7 th Workshop on Asian Translation.We participated in the Indic language multilingual sub-task as well as the NICT-SAP multilingual multi-domain sub-task.We focused on naive many-to-many NMT models which gave reasonable translation quality despite their simplicity.Our observations are twofold: (a.) Many-to-many models suffer from a lack of consistency where the translation quality for some language pairs is very good but for some others it is terrible when compared against one-to-many and many-toone baselines.(b.) Oversampling smaller corpora does not necessarily give the best translation quality for the language pair associated with that pair.

This paper proposes an English-Chinese machine translation research method based on transfer learning. First, it expounds the theory of neural machine translation and transfer learning and related technologies. Neural machine translation is discussed, the advantages and disadvantages of various models are introduced, and the transformer neural machine translation model framework is selected. For low-resource Chinese-English parallel corpus and Tibetan-Chinese parallel corpus, 30 million Chinese-English parallel corpora, 100,000 Chinese-English low-resource parallel corpora, and 100,000 Tibetan-Chinese parallel corpora were used to pretrain the transformer machine translation architecture. The decoders are all composed of 6 identical hidden layers, the initialization of the model parameters is done by the transformer uniform distribution, and the model training uses Adam as the optimizer. In the model transfer part, the parameters with the better effect of the pretrained model are transferred to the low-resource Chinese-English and Tibetan-Chinese machine translation model training, so as to achieve the purpose of knowledge transfer. The results show that the model transfer learning of low-resource Chinese-English parallel corpus improves the translation system's translation by 3.97 BLEU values compared with the translation system without transfer learning at 0.34 BLEU values. Model transfer learning on low-resource Tibetan-Chinese parallel corpus increases the BLEU value by 2.64 BLEU compared to the translation system without transfer learning. The neural machine translation system that uses BPE technology for preprocessing plus model transfer learning is compared to the translation system that only performs transfer learning and shows an improved 0.26 BLEU value.It is verified that the transfer learning method proposed in this paper has a certain improvement in the effect of low-resource Chinese-English and Tibetan-Chinese neural machine translation models.

Machine translation is a computer-based translation process that receives a set of words from a particular humanreadable language as input and outputs a second set of words in the intended human-readable language. Machine translation has assisted linguists and sociologists all across the world. A machine translation model may be statistical, rule-based, or based on neural networks. The Neural Machine Translation (NMT) model was introduced in response to the numerous drawbacks of rule- based and statistical-based machine translation models. This study adds a parallel corpus of Assamese and English and develops the NMT system, a transformer model with a self- attention mechanism. For translation from English to Assamese, the system received a higher bilingual evaluation understudy (BLEU) score of 7.6, and for translation from Assamese to English,a BLEU score of 23.2.

The dominant neural machine translation (NMT) models that based on the encoder-decoder architecture have recently achieved the state-of-the-art performance. Traditionally, the NMT models only depend on the representations learned during training for mapping a source sentence into the target domain. However, the learned representations often suffer from implicit and inadequately informed properties. In this paper, we propose a novel bilingual topic enhanced NMT (BLTNMT) model to improve translation performance by incorporating bilingual topic knowledge into NMT. Specifically, the bilingual topic knowledge is included into the hidden states of both encoder and decoder, as well as the attention mechanism. With this new setting, the proposed BLT-NMT has access to the background knowledge implied in bilingual topics which is beyond the sequential context, and enables the attention mechanism to attend to topic-level attentions for generating accurate target words during translation. Experimental results show that the proposed model consistently outperforms the traditional RNNsearch and the previous topic-informed NMT on Chinese-English and EnglishGerman translation tasks. We also introduce the bilingual topic knowledge into the newly emerged Transformer base model on English-German translation and achieve a notable improvement.

Recent works have shown that Neural Machine Translation (NMT) models achieve impressive performance, however, questions about understanding the behavior of these models remain unanswered. We investigate the unexpected volatility of NMT models where the input is semantically and syntactically correct. We discover that with trivial modifications of source sentences, we can identify cases where unexpected changes happen in the translation and in the worst case lead to mistranslations. This volatile behavior of translating extremely similar sentences in surprisingly different ways highlights the underlying generalization problem of current NMT models. We find that both RNN and Transformer models display volatile behavior in 26% and 19% of sentence variations, respectively.

Machine translation (MT) models usually translate a text at sentence level by considering isolated sentences, which is based on a strict assumption that the sentences in a text are independent of one another.However, the fact is that the texts at discourse level have properties going beyond individual sentences.These properties reveal texts in the frequency and distribution of words, word senses, referential forms and syntactic structures.Disregarding dependencies across sentences will harm translation quality especially in terms of coherence, cohesion, and consistency.To solve these problems, several approaches have previously been investigated for conventional statistical machine translation (SMT).With the fast growth of neural machine translation (NMT), discourse-level NMT has drawn increasing attention from researchers.In this work, we review major works on addressing discourse related problems for both SMT and NMT models with a survey of recent trends in the fields.

To improve the function of machine translation to adapt to global language translation, the work takes deep neural network (DNN) as the basic theory, carries out transfer learning and neural network translation modeling, and optimizes the word alignment function in machine translation performance. First, the work implements a deep learning translation network model for English translation. On this basis, the neural machine translation model is designed under transfer learning. The random shielding method is introduced to implement the language training model, and the machine translation is slightly adjusted as the goal of transfer learning, thereby improving the semantic understanding ability in translation performance. Meanwhile, the work design introduces the method of word alignment optimization and optimizes the performance of word alignment in the transformer system by using word corpus. The experimental results show that the proposed method reduces the average alignment error rate by 8.1%, 24.4%, and 22.1% in EnRo (English-Roman), EnGe (English-German), and EnFr (English-French), respectively, compared with the previous algorithms. Compared with the designed optimization method, the word alignment error rate is lower than that of traditional methods. The modeling and optimization method is feasible, which can effectively solve the problems of insufficient information utilization, large parameter scale, and difficult storage in the process of machine translation. Additionally, it provides a feasible idea and direction for the optimization and improvement in neural machine translation (NMT) system.

Machine translation aims to minimize the language barrier between people of different linguistic backgrounds. Machine translation is an automatic translation technique between pairs of different languages. Machine translation using neural network or neural machine translation came into the picture due to several limitation associated with its predecessors which are rule based and statistical based models. For large data sets with a rich range of vocabulary, the neural network machine translation system provides fair translation accuracy. We have observed that there remain very few machine translators dedicated to Indian languages, especially those spoken in the North-East area. So our paper mainly focuses on implementing a neural machine translation system for the English-Assamese language. In this paper we used five different neural machine translation models for English-Assamese language pair using LSTMs and GRUs, along with an attention layer. We have made a comparison analysis based on the performance results of these models. We used BLEU Score to calculate the accuracy of these five models, thus achieving a BLEU score of 34.168%, which was the highest among the five models.

The hypotactic structural relation between clauses plays an important role in improving the discourse coherence of document-level translation. However, the standard neural machine translation (NMT) models do not explicitly model the hypotactic relationship between clauses, which usually leads to structurally incorrect translations of long and complex sentences. This problem is particularly noticeable on Chinese-to-English translation task of complex sentences due to the grammatical form distinction between English and Chinese. English is rich in grammatical form (e.g. verb morphological changes and subordinating conjunctions) while Chinese is poor in grammatical form. These linguistic phenomena make it a challenge for NMT to learn the hypotactic structure knowledge from Chinese as well as the structure alignment between Chinese and English. To address these issues, we propose to model the hypotactic structure for Chinese-to-English complex sentence translation by introducing hypotactic structure knowledge. Specifically, we annotate and build a hypotactic structure aligned parallel corpus that provides rich hypotactic structure knowledge for NMT. Moreover, we further propose a structure-infused neural framework to combine the hypotactic structure knowledge with the NMT model through two integrating strategies. In particular, we introduce a specific structure-aware loss to encourage the NMT model to better learn the structure knowledge. Experimental results on WMT17, WMT18 and WMT19 Chinese-to-English translation tasks demonstrate the effectiveness of the proposed methods.

Classifiers in natural language processing (NLP) often have a large number of output classes. For example, neural language models (LMs) and machine translation (MT) models both predict tokens from a vocabulary of thousands. The Softmax output layer of these models typically receives as input a dense feature representation, which has much lower dimensionality than the output. In theory, the result is some words may be impossible to be predicted via argmax, irrespective of input features, and empirically, there is evidence this happens in small language models (Demeter et al., 2020). In this paper we ask whether it can happen in practical large language models and translation models. To do so, we develop algorithms to detect such unargmaxable tokens in public models. We find that 13 out of 150 models do indeed have such tokens; however, they are very infrequent and unlikely to impact model quality. We release our algorithms and code so that others can test their models.

Zero pronouns (ZPs) are frequently omitted in pro-drop languages, but should be recalled in non-pro-drop languages. This discourse phenomenon poses a significant challenge for machine translation (MT) when translating texts from pro-drop to non-pro-drop languages. In this paper, we propose a unified and discourse-aware ZP translation approach for neural MT models. Specifically, we jointly learn to predict and translate ZPs in an end-to-end manner, allowing both components to interact with each other. In addition, we employ hierarchical neural networks to exploit discourse-level context, which is beneficial for ZP prediction and thus translation. Experimental results on both Chinese-English and Japanese-English data show that our approach significantly and accumulatively improves both translation performance and ZP prediction accuracy over not only baseline but also previous works using external ZP prediction models. Extensive analyses confirm that the performance improvement comes from the alleviation of different kinds of errors especially caused by subjective ZPs.

In natural Language Processing, Machine Translation is still challenging for low resource regional languages like Hindi and Punjabi because of morphology, script variations and Semantic Ambiguity. Previous research on Hindi- Punjabi translation has mainly focused on distinct methodologies like dictionary-based machine translation, rule-based translation, statistical machine translation and pure neural machine translation. This research proposes a cascade Hybrid MT framework for Hindi-Punjabi Translation integrating Dictionary-based direct mapping for known words, Example based machine translation (EBMT) for similar sentence patterns and Neural Machine Translation for complex linguistic sentences. The proposed system optimizes computational efficiency through knowledge-driven pipelines that prioritize low-cost translation methods before invoking expensive neural networks, therefore proposed hybrid machine translation framework is significantly more computationally efficient than pure Neural Machine Translation (NMT) models. Experimental results demonstrate that the hybrid approach achieves superior performance compared to individual methods, with BLEU scores of 42.5 for dictionary matches, 38.2 for EBMT matches, and 35.7 for NMT translations.

Neural Machine Translation (NMT) models often lack diversity in their generated translations, even when paired with search algorithm, like beam search. A challenge is that the diversity in translations are caused by the variability in the target language, and cannot be inferred from the source sentence alone. In this paper, we propose to explicitly model this one-to-many mapping by conditioning the decoder of a NMT model on a latent variable that represents the domain of target sentences. The domain is a discrete variable generated by a target encoder that is jointly trained with the NMT model. The predicted domain of target sentences are given as input to the decoder during training. At inference, we can generate diverse translations by decoding with different domains. Unlike our strongest baseline (Shen et al., 2019), our method can scale to any number of domains without affecting the performance or the training time. We assess the quality and diversity of translations generated by our model with several metrics, on three different datasets.

Sanskrit is one of the earliest native languages and is correctly described as "the gods' language" because of its wide use in Indian religious literature from the past. However, it is becoming less popular in modern India. Due in significant part to the need for more materials for translation both in and out of Sanskrit, it is no longer commonly utilized. This study explores the feasibility of using machine translation (MT) to provide a link between Sanskrit and, one of the earliest native languages, and its contemporary descendant Hindi. A study was conducted between existing modelling methodologies, notably Statistical machine translation (SMT), and the proposed novel deep learning-based Machine translation strategy using a manually created parallel corpus for the Sanskrit-Hindi language pair. While SMT creates interpretations by mapping phrases from the languages of the source and destination, statistical models, and bilingual text corpora for learning parameters, neural machine translation (NMT) frequently models entire phrases in a single integrated model, using a convolutional neural network to calculate the probability of a word sequence. The proposed NMT model is implemented using an encoder-decoder with an attention mechanism paradigm and the inclusion of gated recurrent units. Our approach involved development of a novel model for Sanskrit-Hindi machine translation using deep learning and the creation of parallel corpora for the Sanskrit-Hindi language pair. The proposed model is evaluated on automated and human-based metrics, and results show that our proposed deep learning-based model outperforms statistical modelling techniques on Moses, surpassing them both with a BLEU score of 53.8% compared to 34.56%. This article examines the undiscovered area of machine translation from Sanskrit to Hindi and discusses the main benefits and drawbacks of statistical and neural machine translation while providing a fresh viewpoint on the subject.