Neural machine translation (MT) models obtain state-of-the-art performance while maintaining a simple, end-to-end architecture. However, little is known about what these models learn about source and target languages during the training process. In this work, we analyze the representations learned by neural MT models at various levels of granularity and empirically evaluate the quality of the representations for learning morphology through extrinsic part-of-speech and morphological tagging tasks. We conduct a thorough investigation along several parameters: word-based vs. character-based representations, depth of the encoding layer, the identity of the target language, and encoder vs. decoder representations. Our data-driven, quantitative evaluation sheds light on important aspects in the neural MT system and its ability to capture word structure.
Center for Information and Language Processing（信息与语言处理中心）；ILLC, University of Amsterdam（阿姆斯特丹大学逻辑、语言与计算研究所）；CIS, LMU Munich（慕尼黑大学信息与语言处理中心）；LMU Munich（慕尼黑大学）
Morphologically rich languages pose difficulties to machine translation. Machine translation engines that rely on statistical learning from parallel training data, such as state-of-the-art neural systems, face challenges especially with rich morphology on the output language side. Key challenges of rich target-side morphology in data-driven machine translation include: (1) A large amount of differently inflected word surface forms entails a larger vocabulary and thus data sparsity. (2) Some inflected forms of infrequent terms typically do not appear in the training corpus, which makes closed-v…
Neural Machine Translation (NMT) in low-resource settings and of morphologically rich languages is made difficult in part by data sparsity of vocabulary words. Several methods have been used to help reduce this sparsity, notably Byte-Pair Encoding (BPE) and a character-based CNN layer (charCNN). However, the charCNN has largely been neglected, possibly because it has only been compared to BPE rather than combined with it. We argue for a reconsideration of the charCNN, based on cross-lingual improvements on low-resource data. We translate from 8 languages into English, using a multi-way paralle…
University of Trento（特伦托大学）；University of Edinburgh（爱丁堡大学）；University of Zürich（苏黎世大学）；University of Amsterdam（阿姆斯特丹大学）
Translation into morphologically-rich languages challenges neural machine translation (NMT) models with extremely sparse vocabularies where atomic treatment of surface forms is unrealistic. This problem is typically addressed by either pre-processing words into subword units or performing translation directly at the level of characters. The former is based on word segmentation algorithms optimized using corpus-level statistics with no regard to the translation task. The latter learns directly from translation data but requires rather deep architectures. In this paper, we propose to translate w…
Indian Institute of Technology (IIT) Bombay（印度理工学院孟买分校）
When translating into morphologically rich languages, Statistical MT approaches face the problem of data sparsity. The severity of the sparseness problem will be high when the corpus size of morphologically richer language is less. Even though we can use factored models to correctly generate morphological forms of words, the problem of data sparseness limits their performance. In this paper, we describe a simple and effective solution which is based on enriching the input corpora with various morphological forms of words. We use this method with the phrase-based and factor-based experiments on…
TALP Research Center（TALP研究中心）；Universitat Politècnica de Catalunya（加泰罗尼亚理工大学）
Morphology in unbalanced languages remains a big challenge in the context of machine translation. In this paper, we propose to de-couple machine translation from morphology generation in order to better deal with the problem. We investigate the morphology simplification with a reasonable trade-off between expected gain and generation complexity. For the Chinese-Spanish task, optimum morphological simplification is in gender and number. For this purpose, we design a new classification architecture which, compared to other standard machine learning techniques, obtains the best results. This prop…
Phrase-based Statistical models are more commonly used as they perform optimally in terms of both, translation quality and complexity of the system. Hindi and in general all Indian languages are morphologically richer than English. Hence, even though Phrase-based systems perform very well for the less divergent language pairs, for English to Indian language translation, we need more linguistic information (such as morphology, parse tree, parts of speech tags, etc.) on the source side. Factored models seem to be useful in this case, as Factored models consider word as a vector of factors. These…
Columbia University（哥伦比亚大学）；New York University Abu Dhabi（纽约大学阿布扎比分校）
The lack of parallel data for many language pairs is an important challenge to statistical machine translation (SMT). One common solution is to pivot through a third language for which there exist parallel corpora with the source and target languages. Although pivoting is a robust technique, it introduces some low quality translations especially when a poor morphology language is used as the pivot between rich morphology languages. In this paper, we examine the use of synchronous morphology constraint features to improve the quality of phrase pivot SMT. We compare hand-crafted constraints to t…
This paper presents a scalable method for integrating compositional morphological representations into a vector-based probabilistic language model. Our approach is evaluated in the context of log-bilinear language models, rendered suitably efficient for implementation inside a machine translation decoder by factoring the vocabulary. We perform both intrinsic and extrinsic evaluations, presenting results on a range of languages which demonstrate that our model learns morphological representations that both perform well on word similarity tasks and lead to substantial reductions in perplexity. W…
University of Zurich（苏黎世大学）；University of Edinburgh（爱丁堡大学）
Data-driven subword segmentation has become the default strategy for open-vocabulary machine translation and other NLP tasks, but may not be sufficiently generic for optimal learning of non-concatenative morphology. We design a test suite to evaluate segmentation strategies on different types of morphological phenomena in a controlled, semi-synthetic setting. In our experiments, we compare how well machine translation models trained on subword- and character-level can translate these morphological phenomena. We find that learning to analyse and generate morphologically complex surface represen…
Informatics Institute, University of Amsterdam（阿姆斯特丹大学信息学研究所）
Attention in neural machine translation provides the possibility to encode relevant parts of the source sentence at each translation step. As a result, attention is considered to be an alignment model as well. However, there is no work that specifically studies attention and provides analysis of what is being learned by attention models. Thus, the question still remains that how attention is similar or different from the traditional alignment. In this paper, we provide detailed analysis of attention and compare it to traditional alignment. We answer the question of whether attention is only ca…
Xinjiang Technical Institute of Physics & Chemistry, Chinese Academy of Sciences（中国科学院新疆理化技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Neural machine translation (NMT) has achieved impressive performance on machine translation task in recent years. However, in consideration of efficiency, a limited-size vocabulary that only contains the top-N highest frequency words are employed for model training, which leads to many rare and unknown words. It is rather difficult when translating from the low-resource and morphologically-rich agglutinative languages, which have complex morphology and large vocabulary. In this paper, we propose a morphological word segmentation method on the source-side for NMT that incorporates morphology kn…
TextShuttle AG（TextShuttle股份公司）；University of Zurich（苏黎世大学）
Machine translation (MT) has been shown to produce a number of errors that require human post-editing, but the extent to which professional human translation (HT) contains such errors has not yet been compared to MT. We compile pre-translated documents in which MT and HT are interleaved, and ask professional translators to flag errors and post-edit these documents in a blind evaluation. We find that the post-editing effort for MT segments is only higher in two out of three language pairs, and that the number of segments with wrong terminology, omissions, and typographical problems is similar i…
Machine Translation is one of the major oldest and the most active research area in Natural Language Processing. Currently, Statistical Machine Translation (SMT) dominates the Machine Translation research. Statistical Machine Translation is an approach to Machine Translation which uses models to learn translation patterns directly from data, and generalize them to translate a new unseen text. The SMT approach is largely language independent, i.e. the models can be applied to any language pair. Statistical Machine Translation (SMT) attempts to generate translations using statistical methods bas…
University of Groningen（格罗宁根大学）；Dublin City University（都柏林城市大学）
Given the rise of a new approach to MT, Neural MT (NMT), and its promising performance on different text types, we assess the translation quality it can attain on what is perceived to be the greatest challenge for MT: literary text. Specifically, we target novels, arguably the most popular type of literary text. We build a literary-adapted NMT system for the English-to-Catalan translation direction and evaluate it against a system pertaining to the previous dominant paradigm in MT: statistical phrase-based MT (PBSMT). To this end, for the first time we train MT systems, both NMT and PBSMT, on …
Insight Centre for Data Analytics, National University of Ireland Galway（爱尔兰国立大学戈尔韦分校Insight数据分析中心）；University of Kurdistan Hewlêr（库尔德斯坦埃尔比勒大学）
Morphological analysis is the study of the formation and structure of words. It plays a crucial role in various tasks in Natural Language Processing (NLP) and Computational Linguistics (CL) such as machine translation and text and speech generation. Kurdish is a less-resourced multi-dialect Indo-European language with highly inflectional morphology. In this paper, as the first attempt of its kind, the morphology of the Kurdish language (Sorani dialect) is described from a computational point of view. We extract morphological rules which are transformed into finite-state transducers for generat…
Informatics Institute（信息学院）；University of Amsterdam（阿姆斯特丹大学）；Department of Computer Science（计算机科学系）；University of Washington（华盛顿大学）
Machine translation systems require semantic knowledge and grammatical understanding. Neural machine translation (NMT) systems often assume this information is captured by an attention mechanism and a decoder that ensures fluency. Recent work has shown that incorporating explicit syntax alleviates the burden of modeling both types of knowledge. However, requiring parses is expensive and does not explore the question of what syntax a model needs during translation. To address both of these issues we introduce a model that simultaneously translates while inducing dependency trees. In this way, w…
Soochow University（苏州大学）；Alibaba Group（阿里巴巴集团）；Singapore University of Technology and Design（新加坡科技设计大学）
Neural machine translation (NMT) suffers a performance deficiency when a limited vocabulary fails to cover the source or target side adequately, which happens frequently when dealing with morphologically rich languages. To address this problem, previous work focused on adjusting translation granularity or expanding the vocabulary size. However, morphological information is relatively under-considered in NMT architectures, which may further improve translation quality. We propose a novel method, which can not only reduce data sparsity but also model morphology through a simple but effective mec…
Nanjing University（南京大学）；Toutiao AI Lab（字节跳动人工智能实验室）；University of Waterloo（滑铁卢大学）；Tencent AI Lab（腾讯人工智能实验室）
Existing neural machine translation systems do not explicitly model what has been translated and what has not during the decoding phase. To address this problem, we propose a novel mechanism that separates the source information into two parts: translated Past contents and untranslated Future contents, which are modeled by two additional recurrent layers. The Past and Future contents are fed to both the attention model and the decoder states, which offers NMT systems the knowledge of translated and untranslated contents. Experimental results show that the proposed approach significantly improv…
While recent advances in deep learning led to significant improvements in machine translation, neural machine translation is often still not able to continuously adapt to the environment. For humans, as well as for machine translation, bilingual dictionaries are a promising knowledge source to continuously integrate new knowledge. However, their exploitation poses several challenges: The system needs to be able to perform one-shot learning as well as model the morphology of source and target language. In this work, we proposed an evaluation framework to assess the ability of neural machine tra…
The University of Melbourne（墨尔本大学）；Andalas University（安达拉斯大学）
Although some linguists (Rusmali et al., 1985; Crouch, 2009) have fairly attempted to define the morphology and syntax of Minangkabau, information processing in this language is still absent due to the scarcity of the annotated resource. In this work, we release two Minangkabau corpora: sentiment analysis and machine translation that are harvested and constructed from Twitter and Wikipedia. We conduct the first computational linguistics in Minangkabau language employing classic machine learning and sequence-to-sequence models such as LSTM and Transformer. Our first experiments show that the cl…
Machine translation (MT) involving Indigenous languages, including those possibly endangered, is challenging due to lack of sufficient parallel data. We describe an approach exploiting bilingual and multilingual pretrained MT models in a transfer learning setting to translate from Spanish to ten South American Indigenous languages. Our models set new SOTA on five out of the ten language pairs we consider, even doubling performance on one of these five pairs. Unlike previous SOTA that perform data augmentation to enlarge the train sets, we retain the low-resource setting to test the effectivene…
In machine translation it is common phenomenon that machine-readable dictionaries and standard parsing rules are not enough to ensure accuracy in parsing and translating English phrases into Korean language, which is revealed in misleading translation results due to consequent structural and semantic ambiguities. This paper aims to suggest a solution to structural and semantic ambiguities due to the idiomaticity and non-grammaticalness of phrases commonly used in English language by applying bilingual phrase database in English-Korean Machine Translation (EKMT). This paper firstly clarifies wh…
New York University Abu Dhabi（纽约大学阿布扎比分校）；Ohio State University（俄亥俄州立大学）；Carnegie Mellon University Qatar（卡内基梅隆大学卡塔尔分校）
We present the second ever evaluated Arabic dialect-to-dialect machine translation effort, and the first to leverage external resources beyond a small parallel corpus. The subject has not previously received serious attention due to lack of naturally occurring parallel data; yet its importance is evidenced by dialectal Arabic's wide usage and breadth of inter-dialect variation, comparable to that of Romance languages. Our results suggest that modeling morphology and syntax significantly improves dialect-to-dialect translation, though optimizing such data-sparse models requires consideration of…
University of Edinburgh（爱丁堡大学）；New York University（纽约大学）；Aalborg University（奥尔堡大学）
Theoretical work in morphological typology offers the possibility of measuring morphological diversity on a continuous scale. However, literature in Natural Language Processing (NLP) typically labels a whole language with a strict type of morphology, e.g. fusional or agglutinative. In this work, we propose to reduce the rigidity of such claims, by quantifying morphological typology at the word and segment level. We consider Payne (2017)'s approach to classify morphology using two indices: synthesis (e.g. analytic to polysynthetic) and fusion (agglutinative to fusional). For computing synthesis…
FBK（FBK（布鲁诺·凯斯勒基金会））；University of Amsterdam（阿姆斯特丹大学）
Within the field of Statistical Machine Translation (SMT), the neural approach (NMT) has recently emerged as the first technology able to challenge the long-standing dominance of phrase-based approaches (PBMT). In particular, at the IWSLT 2015 evaluation campaign, NMT outperformed well established state-of-the-art PBMT systems on English-German, a language pair known to be particularly hard because of morphology and syntactic differences. To understand in what respects NMT provides better translation quality than PBMT, we perform a detailed analysis of neural versus phrase-based SMT outputs, l…
Neural machine translation (NMT) systems operate primarily on words (or sub-words), ignoring lower-level patterns of morphology. We present a character-aware decoder designed to capture such patterns when translating into morphologically rich languages. We achieve character-awareness by augmenting both the softmax and embedding layers of an attention-based encoder-decoder model with convolutional neural networks that operate on the spelling of a word. To investigate performance on a wide variety of morphological phenomena, we translate English into 14 typologically diverse target languages usi…
University of Trento（特伦托大学）；Fondazione Bruno Kessler（布鲁诺·凯斯勒基金会）；MMT Srl（MMT有限公司）
Recently, neural machine translation (NMT) has been extended to multilinguality, that is to handle more than one translation direction with a single system. Multilingual NMT showed competitive performance against pure bilingual systems. Notably, in low-resource settings, it proved to work effectively and efficiently, thanks to shared representation space that is forced across languages and induces a sort of transfer-learning. Furthermore, multilingual NMT enables so-called zero-shot inference across language pairs never seen at training time. Despite the increasing interest in this framework, …
This article overviews the current state of the English-Lithuanian-English machine translation system. The first part of the article describes the problems that system poses today and what actions will be taken to solve them in the future. The second part of the article tackles the main issue of the translation process. Article briefly overviews the word sense disambiguation for MT technique using Google.
Sarcasm is a form of speech in which speakers say the opposite of what they truly mean in order to convey a strong sentiment. In other words, "Sarcasm is the giant chasm between what I say, and the person who doesn't get it.". In this paper we present the novel task of sarcasm interpretation, defined as the generation of a non-sarcastic utterance conveying the same message as the original sarcastic one. We introduce a novel dataset of 3000 sarcastic tweets, each interpreted by five human judges. Addressing the task as monolingual machine translation (MT), we experiment with MT algorithms and e…
Transformer based models are the modern work horses for neural machine translation (NMT), reaching state of the art across several benchmarks. Despite their impressive accuracy, we observe a systemic and rudimentary class of errors made by transformer based models with regards to translating from a language that doesn't mark gender on nouns into others that do. We find that even when the surrounding context provides unambiguous evidence of the appropriate grammatical gender marking, no transformer based model we tested was able to accurately gender occupation nouns systematically. We release a…
Conditional masked language model (CMLM) training has proven successful for non-autoregressive and semi-autoregressive sequence generation tasks, such as machine translation. Given a trained CMLM, however, it is not clear what the best inference strategy is. We formulate masked inference as a factorization of conditional probabilities of partial sequences, show that this does not harm performance, and investigate a number of simple heuristics motivated by this perspective. We identify a thresholding strategy that has advantages over the standard "mask-predict" algorithm, and provide analyses o…
xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detection
Instituto de Telecomunicações（电信研究所）；CentraleSupélec（巴黎中央理工-高等电力学院）；Université Paris-Saclay（巴黎萨克雷大学）；Instituto Superior Técnico（高等理工学院）；University of Lisbon（里斯本大学）；Unbabel（Unbabel公司）；INESC-ID（葡萄牙系统与计算机工程研究所）
Tianjin University（天津大学）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
In the era of deep learning, word embeddings are essential when dealing with text tasks. However, storing and accessing these embeddings requires a large amount of space. This is not conducive to the deployment of these models on resource-limited devices. Combining the powerful compression capability of tensor products, we propose a word embedding compression method with morphological augmentation, Morphologically-enhanced Tensorized Embeddings (MorphTE). A word consists of one or more morphemes, the smallest units that bear meaning or have a grammatical function. MorphTE represents a word emb…
This paper describes the University of Maryland's submission to the Special Task on Formality Control for Spoken Language Translation at \iwslt, which evaluates translation from English into 6 languages with diverse grammatical formality markers. We investigate to what extent this problem can be addressed with a \textit{single multilingual model}, simultaneously controlling its output for target language and formality. Results show that this strategy can approach the translation quality and formality control achieved by dedicated translation models. However, the nature of the underlying pre-tr…
ADAPT Centre（ADAPT研究中心）；School of Computing（计算学院）；Dublin City University（都柏林城市大学）
Recently, neural machine translation (NMT) has emerged as a powerful alternative to conventional statistical approaches. However, its performance drops considerably in the presence of morphologically rich languages (MRLs). Neural engines usually fail to tackle the large vocabulary and high out-of-vocabulary (OOV) word rate of MRLs. Therefore, it is not suitable to exploit existing word-based models to translate this set of languages. In this paper, we propose an extension to the state-of-the-art model of Chung et al. (2016), which works at the character level and boosts the decoder with target…
The state of the art of handling rich morphology in neural machine translation (NMT) is to break word forms into subword units, so that the overall vocabulary size of these units fits the practical limits given by the NMT model and GPU memory capacity. In this paper, we compare two common but linguistically uninformed methods of subword construction (BPE and STE, the method implemented in Tensor2Tensor toolkit) and two linguistically-motivated methods: Morfessor and one novel method, based on a derivational dictionary. Our experiments with German-to-Czech translation, both morphologically rich…
Massachusetts Institute of Technology（麻省理工学院）；University of Washington（华盛顿大学）
Character-based neural machine translation (NMT) models alleviate out-of-vocabulary issues, learn morphology, and move us closer to completely end-to-end translation systems. Unfortunately, they are also very brittle and easily falter when presented with noisy data. In this paper, we confront NMT models with synthetic and natural sources of noise. We find that state-of-the-art models fail to translate even moderately noisy texts that humans have no trouble comprehending. We explore two approaches to increase model robustness: structure-invariant word representations and robust training on nois…
Low-resource Machine Translation: what for? who for? An observational study on a dedicated Tetun language translation service
School of Computing and Information Systems, The University of Melbourne（墨尔本大学计算与信息学院）；Instituto Nacional de Linguística, Dili（迪利国家语言研究所）；The Australian National University（澳大利亚国立大学）
ADAPT Centre（ADAPT中心）；School of Computing（计算学院）；National College of Ireland（爱尔兰国家学院）；Dublin City University（都柏林城市大学）
Research on Machine Translation (MT) has achieved important breakthroughs in several areas. While there is much more to be done in order to build on this success, we believe that the language industry needs better ways to take full advantage of current achievements. Due to a combination of factors, including time, resources, and skills, businesses tend to apply pragmatism into their AI workflows. Hence, they concentrate more on outcomes, e.g. delivery, shipping, releases, and features, and adopt high-level working production solutions, where possible. Among the features thought to be helpful f…
Facebook Applied Machine Learning（Facebook应用机器学习）；Facebook AI Research（Facebook人工智能研究院）；Sorbonne Universités（索邦大学）；Johns Hopkins University（约翰斯·霍普金斯大学）
For machine translation, a vast majority of language pairs in the world are considered low-resource because they have little parallel data available. Besides the technical challenges of learning with limited supervision, it is difficult to evaluate methods trained on low-resource language pairs because of the lack of freely and publicly available benchmarks. In this work, we introduce the FLoRes evaluation datasets for Nepali-English and Sinhala-English, based on sentences translated from Wikipedia. Compared to English, these are languages with very different morphology and syntax, for which l…
Yandex；University of Amsterdam（阿姆斯特丹大学）；University of Edinburgh（爱丁堡大学）；University of Zurich（苏黎世大学）
Standard machine translation systems process sentences in isolation and hence ignore extra-sentential information, even though extended context can both prevent mistakes in ambiguous cases and improve translation coherence. We introduce a context-aware neural machine translation model designed in such way that the flow of information from the extended context to the translation model can be controlled and analyzed. We experiment with an English-Russian subtitles dataset, and observe that much of what is captured by our model deals with improving pronoun translation. We measure correspondences …
National Research Council Canada（加拿大国家研究委员会）；Google（谷歌）
Neural machine translation represents an exciting leap forward in translation quality. But what longstanding weaknesses does it resolve, and which remain? We address these questions with a challenge set approach to translation evaluation and error analysis. A challenge set consists of a small set of sentences, each hand-designed to probe a system's capacity to bridge a particular structural divergence between languages. To exemplify this approach, we present an English-French challenge set, and use it to analyze phrase-based and neural systems. The resulting analysis provides not only a more f…
Interpreters facilitate multi-lingual meetings but the affordable set of languages is often smaller than what is needed. Automatic simultaneous speech translation can extend the set of provided languages. We investigate if such an automatic system should rather follow the original speaker, or an interpreter to achieve better translation quality at the cost of increased delay. To answer the question, we release Europarl Simultaneous Interpreting Corpus (ESIC), 10 hours of recordings and transcripts of European Parliament speeches in English, with simultaneous interpreting into Czech and German.…
Intelligent selection of training data has proven a successful technique to simultaneously increase training efficiency and translation performance for phrase-based machine translation (PBMT). With the recent increase in popularity of neural machine translation (NMT), we explore in this paper to what extent and how NMT can also benefit from data selection. While state-of-the-art data selection (Axelrod et al., 2011) consistently performs well for PBMT, we show that gains are substantially lower for NMT. Next, we introduce dynamic data selection for NMT, a method in which we vary the selected s…
University of Zurich（苏黎世大学）；University of Edinburgh（爱丁堡大学）
As machine translation (MT) metrics improve their correlation with human judgement every year, it is crucial to understand the limitations of such metrics at the segment level. Specifically, it is important to investigate metric behaviour when facing accuracy errors in MT because these can have dangerous consequences in certain contexts (e.g., legal, medical). We curate ACES, a translation accuracy challenge set, consisting of 68 phenomena ranging from simple perturbations at the word/character level to more complex errors based on discourse and real-world knowledge. We use ACES to evaluate a …
Automatic metrics are fundamental for the development and evaluation of machine translation systems. Judging whether, and to what extent, automatic metrics concur with the gold standard of human evaluation is not a straightforward problem. We show that current methods for judging metrics are highly sensitive to the translations used for assessment, particularly the presence of outliers, which often leads to falsely confident conclusions about a metric's efficacy. Finally, we turn to pairwise system ranking, developing a method for thresholding performance improvement under an automatic metric …
