Sentence-level (SL) machine translation (MT) has reached acceptable quality for many high-resourced languages, but not document-level (DL) MT, which is difficult to 1) train with little amount of DL data; and 2) evaluate, as the main methods and data sets focus on SL evaluation. To address the first issue, we present a document-aligned Japanese-English conversation corpus, including balanced, high-quality business conversation data for tuning and testing. As for the second issue, we manually identify the main areas where SL MT fails to produce adequate translations in lack of context. We then create an evaluation set where these phenomena are annotated to alleviate automatic evaluation of DL systems. We train MT models using our corpus to demonstrate how using context leads to improvements.
For Japanese-to-English translation, zero pronouns in Japanese pose a challenge, since the model needs to infer and produce the corresponding pronoun in the target side of the English sentence. However, although fully resolving zero pronouns often needs discourse context, in some cases, the local context within a sentence gives clues to the inference of the zero pronoun. In this study, we propose a data augmentation method that provides additional training signals for the translation model to learn correlations between local context and zero pronouns. We show that the proposed method significa…
This paper describes in outline a method for translating Japanese temporal expressions into English. We argue that temporal expressions form a special subset of language that is best handled as a special module in machine translation. The paper deals with problems of lexical idiosyncrasy as well as the choice of articles and prepositions within temporal expressions. In addition temporal expressions are considered as parts of larger structures, and the question of whether to translate them as noun phrases or adverbials is addressed.
NTT Communication Science Laboratories, NTT Corporation（日本电信电话株式会社通信科学实验室）
Most current machine translation models are mainly trained with parallel corpora, and their translation accuracy largely depends on the quality and quantity of the corpora. Although there are billions of parallel sentences for a few language pairs, effectively dealing with most language pairs is difficult due to a lack of publicly available parallel corpora. This paper creates a large parallel corpus for English-Japanese, a language pair for which only limited resources are available, compared to such resource-rich languages as English-German. It introduces a new web-based English-Japanese par…
Neural machine translation (NMT) has recently become popular in the field of machine translation. However, NMT suffers from the problem of repeating or missing words in the translation. To address this problem, Tu et al. (2017) proposed an encoder-decoder-reconstructor framework for NMT using back-translation. In this method, they selected the best forward translation model in the same manner as Bahdanau et al. (2015), and then trained a bi-directional translation model as fine-tuning. Their experiments show that it offers significant improvement in BLEU scores in Chinese-English translation t…
Graduate School of Informatics, Kyoto University（京都大学信息学研究科）
In the present study, we propose novel sequence-to-sequence pre-training objectives for low-resource machine translation (NMT): Japanese-specific sequence to sequence (JASS) for language pairs involving Japanese as the source or target language, and English-specific sequence to sequence (ENSS) for language pairs involving English. JASS focuses on masking and reordering Japanese linguistic units known as bunsetsu, whereas ENSS is proposed based on phrase structure masking and reordering tasks. Experiments on ASPEC Japanese--English & Japanese--Chinese, Wikipedia Japanese--Chinese, News English-…
We investigate the behaviour of attention in neural models of visually grounded speech trained on two languages: English and Japanese. Experimental results show that attention focuses on nouns and this behaviour holds true for two very typologically different languages. We also draw parallels between artificial neural attention and human attention and show that neural attention focuses on word endings as it has been theorised for human attention. Finally, we investigate how two visually grounded monolingual models can be used to perform cross-lingual speech-to-speech retrieval. For both langua…
This paper proposes an analysis of classifiers into four major types: UNIT, METRIC, GROUP and SPECIES, based on properties of both Japanese and English. The analysis makes possible a uniform and straightforward treatment of noun phrases headed by classifiers in Japanese-to-English machine translation, and has been implemented in the MT system ALT-J/E. Although the analysis is based on the characteristics of, and differences between, Japanese and English, it is shown to be also applicable to the unrelated language Thai.
Graduate School of Information Science and Technology, Osaka University（大阪大学信息科学与技术研究生院）；Institute for Datability Science, Osaka University（大阪大学数据可用性科学研究所）
The word order between source and target languages significantly influences the translation quality in machine translation. Preordering can effectively address this problem. Previous preordering methods require a manual feature design, making language dependent design costly. In this paper, we propose a preordering method with a recursive neural network that learns features from raw inputs. Experiments show that the proposed method achieves comparable gain in translation quality to the state-of-the-art method but without a manual feature design.
This paper shows the necessity of distinguishing different referential uses of noun phrases in machine translation. We argue that differentiating between the generic, referential and ascriptive uses of noun phrases is the minimum necessary to generate articles and number correctly when translating from Japanese to English. Heuristics for determining these differences are proposed for a Japanese-to-English machine translation system. Finally the results of using the proposed heuristics are shown to have raised the percentage of noun phrases generated with correct use of articles and number in t…
This paper introduces WeChat AI's participation in WMT 2021 shared news translation task on English->Chinese, English->Japanese, Japanese->English and English->German. Our systems are based on the Transformer (Vaswani et al., 2017) with several novel and effective variants. In our experiments, we employ data filtering, large-scale synthetic data generation (i.e., back-translation, knowledge distillation, forward-translation, iterative in-domain knowledge transfer), advanced finetuning approaches, and boosted Self-BLEU based model ensemble. Our constrained systems achieve 36.9, 46.9, 27.8 and 3…
This paper presents a heuristic method that uses information in the Japanese text along with knowledge of English countability and number stored in transfer dictionaries to determine the countability and number of English noun phrases. Incorporating this method into the machine translation system ALT-J/E, helped to raise the percentage of noun phrases generated with correct use of articles and number from 65% to 73%.
Possessive pronouns are used as determiners in English when no equivalent would be used in a Japanese sentence with the same meaning. This paper proposes a heuristic method of generating such possessive pronouns even when there is no equivalent in the Japanese. The method uses information about the use of possessive pronouns in English treated as a lexical property of nouns, in addition to contextual information about noun phrase referentiality and the subject and main verb of the sentence that the noun phrase appears in. The proposed method has been implemented in NTT Communication Science La…
Neural machine translation (NMT) has a drawback in that can generate only high-frequency words owing to the computational costs of the softmax function in the output layer. In Japanese-English NMT, Japanese predicate conjugation causes an increase in vocabulary size. For example, one verb can have as many as 19 surface varieties. In this research, we focus on predicate conjugation for compressing the vocabulary size in Japanese. The vocabulary list is filled with the various forms of verbs. We propose methods using predicate conjugation information without discarding linguistic information. Th…
Nara Institute of Science and Technology（奈良先端科学技术大学院大学）
We present the results of research with the goal of automatically creating a multilingual thesaurus based on the freely available resources of Wikipedia and WordNet. Our goal is to increase resources for natural language processing tasks such as machine translation targeting the Japanese-Spanish language pair. Given the scarcity of resources, we use existing English resources as a pivot for creating a trilingual Japanese-Spanish-English thesaurus. Our approach consists of extracting the translation tuples from Wikipedia, disambiguating them by mapping them to WordNet word senses. We present re…
When translating Japanese nouns into English, we face the problem of articles and numbers which the Japanese language does not have, but which are necessary for the English composition. To solve this difficult problem we classified the referential property and the number of nouns into three types respectively. This paper shows that the referential property and the number of nouns in a sentence can be estimated fairly reliably by the words in the sentence. Many rules for the estimation were written in forms similar to rewriting rules in expert systems. We obtained the correct recognition scores…
University of Georgia（佐治亚大学）；Kyoto University（京都大学）；Osaka University（大阪大学）
Multimodal neural machine translation (NMT) has become an increasingly important area of research over the years because additional modalities, such as image data, can provide more context to textual data. Furthermore, the viability of training multimodal NMT models without a large parallel corpus continues to be investigated due to low availability of parallel sentences with images, particularly for English-Japanese data. However, this void can be filled with comparable sentences that contain bilingual terms and parallel phrases, which are naturally created through media such as social networ…
Verbs are sometimes omitted in Japanese sentences. It is necessary to recover omitted verbs for purposes of language understanding, machine translation, and conversational processing. This paper describes a practical way to recover omitted verbs by using surface expressions and examples. We experimented the resolution of verb ellipses by using this information, and obtained a recall rate of 73% and a precision rate of 66% on test sentences.
Nara Institute of Science and Technology (NAIST)（奈良先端科学技术大学院大学）；RIKEN（日本理化学研究所）
Simultaneous translation is a task in which translation begins before the speaker has finished speaking, so it is important to decide when to start the translation process. However, deciding whether to read more input words or start to translate is difficult for language pairs with different word orders such as English and Japanese. Motivated by the concept of pre-reordering, we propose a couple of simple decision rules using the label of the next constituent predicted by incremental constituent label prediction. In experiments on English-to-Japanese simultaneous translation, the proposed meth…
Anaphora resolution is one of the major problems in natural language processing. It is also one of the important tasks in machine translation and man/machine dialogue. We solve the problem by using surface expressions and examples. Surface expressions are the words in sentences which provide clues for anaphora resolution. Examples are linguistic data which are actually used in conversations and texts. The method using surface expressions and examples is a practical method. This thesis handles almost all kinds of anaphora: i. The referential property and number of a noun phrase ii. Noun phrase …
The referential properties of noun phrases in the Japanese language, which has no articles, are useful for article generation in Japanese-English machine translation and for anaphora resolution in Japanese noun phrases. They are generally classified as generic noun phrases, definite noun phrases, and indefinite noun phrases. In the previous work, referential properties were estimated by developing rules that used clue words. If two or more rules were in conflict with each other, the category having the maximum total score given by the rules was selected as the desired category. The score given…
While the progress of machine translation of written text has come far in the past several years thanks to the increasing availability of parallel corpora and corpora-based training technologies, automatic translation of spoken text and dialogues remains challenging even for modern systems. In this paper, we aim to boost the machine translation quality of conversational texts by introducing a newly constructed Japanese-English business conversation parallel corpus. A detailed analysis of the corpus is provided along with challenging examples for automatic translation. We also experiment with a…
University of Tsukuba（筑波大学）；Japan Patent Information Organization（日本专利信息机构）
Neural machine translation (NMT), a new approach to machine translation, has achieved promising results comparable to those of traditional approaches such as statistical machine translation (SMT). Despite its recent success, NMT cannot handle a larger vocabulary because the training complexity and decoding complexity proportionally increase with the number of target words. This problem becomes even more serious when translating patent documents, which contain many technical terms that are observed infrequently. In this paper, we propose to select phrases that contain out-of-vocabulary words us…
Westlake University（西湖大学）；Sichuan Lan-bridge Information Technology Co., Ltd.（四川蓝桥信息技术有限公司）；Zhejiang University（浙江大学）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Michigan, Ann Arbor（密歇根大学安娜堡分校）
We propose the shared task of cross-lingual conversation summarization, \emph{ConvSumX Challenge}, opening new avenues for researchers to investigate solutions that integrate conversation summarization and machine translation. This task can be particularly useful due to the emergence of online meetings and conferences. We construct a new benchmark, covering 2 real-world scenarios and 3 language directions, including a low-resource language. We hope that \emph{ConvSumX} can motivate researches to go beyond English and break the barrier for non-English speakers to benefit from recent advances of…
Tohoku University（东北大学）；Langsmith Inc.（Langsmith公司）；University of Tokyo（东京大学）；RIKEN（理化学研究所）；NINJAL（国立国语研究所）
In computational psycholinguistics, various language models have been evaluated against human reading behavior (e.g., eye movement) to build human-like computational models. However, most previous efforts have focused almost exclusively on English, despite the recent trend towards linguistic universal within the general community. In order to fill the gap, this paper investigates whether the established results in computational psycholinguistics can be generalized across languages. Specifically, we re-examine an established generalization -- the lower perplexity a language model has, the more …
Kyoto University（京都大学）；National Institute of Information and Communications Technology（国立信息与通信技术研究所）
Neural machine translation (NMT) needs large parallel corpora for state-of-the-art translation quality. Low-resource NMT is typically addressed by transfer learning which leverages large monolingual or parallel corpora for pre-training. Monolingual pre-training approaches such as MASS (MAsked Sequence to Sequence) are extremely effective in boosting NMT quality for languages with small parallel corpora. However, they do not account for linguistic information obtained using syntactic analyzers which is known to be invaluable for several Natural Language Processing (NLP) tasks. To this end, we p…
University of Copenhagen（哥本哈根大学）；Tokyo Institute of Technology（东京工业大学）
Most neural machine translation models only rely on pairs of parallel sentences, assuming syntactic information is automatically learned by an attention mechanism. In this work, we investigate different approaches to incorporate syntactic knowledge in the Transformer model and also propose a novel, parameter-free, dependency-aware self-attention mechanism that improves its translation quality, especially for long sentences and in low-resource scenarios. We show the efficacy of each approach on WMT English-German and English-Turkish, and WAT English-Japanese translation tasks.
Recently, several types of Japanese-to-English machine translation systems have been developed, but all of them require an initial process of rewriting the original text into easily translatable Japanese. Therefore these systems are unsuitable for translating information that needs to be speedily disseminated. To overcome this limitation, a Multi-Level Translation Method based on the Constructive Process Theory has been proposed. This paper describes the benefits of using this method in the Japanese-to-English machine translation system ALT-J/E. In comparison with conventional compositional me…
Interpretable Cross-Lingual Alignment in Small Language Models: Probing Cultural and Pragmatic Reasoning in Japanese-English Bilingual LLMs
Sakaguchi–Inui Laboratory (Tohoku University)（东北大学坂口–乾实验室）；Natural Language Understanding Team at RIKEN AIP（理化学研究所先进智能项目中心自然语言理解团队）；Swallow Project at the Institute of Science Tokyo（东京科学大学Swallow项目）；Sakana AI；Tohoku University（东北大学）；RIKEN AIP（理化学研究所先进智能项目中心）；Institute of Science Tokyo（东京科学大学）
Stanford University（斯坦福大学）；Rakuten Institute of Technology（乐天技术研究所）；Google Brain（谷歌大脑）
In this paper we describe the Japanese-English Subtitle Corpus (JESC). JESC is a large Japanese-English parallel corpus covering the underrepresented domain of conversational dialogue. It consists of more than 3.2 million examples, making it the largest freely available dataset of its kind. The corpus was assembled by crawling and aligning subtitles found on the web. The assembly process incorporates a number of novel preprocessing elements to ensure high monolingual fluency and accurate bilingual alignments. We summarize its contents and evaluate its quality using human experts and baseline m…
This paper applies an existing query translation method to cross-language patent retrieval. In our method, multiple dictionaries are used to derive all possible translations for an input query, and collocational statistics are used to resolve translation ambiguity. We used Japanese/English parallel patent abstracts to perform comparative experiments, where our method outperformed a simple dictionary-based query translation method, and achieved 76% of monolingual retrieval in terms of average precision.
Kyoto University（京都大学）；National Institute of Information and Communications Technology（信息通信研究机构）
Sequence-to-sequence (S2S) pre-training using large monolingual data is known to improve performance for various S2S NLP tasks in low-resource settings. However, large monolingual corpora might not always be available for the languages of interest (LOI). To this end, we propose to exploit monolingual corpora of other languages to complement the scarcity of monolingual corpora for the LOI. A case study of low-resource Japanese-English neural machine translation (NMT) reveals that leveraging large Chinese and French monolingual corpora can help overcome the shortage of Japanese and English monol…
This paper describes experiments carried out using a variety of machine-learning methods, including the k-nearest neighborhood method that was used in a previous study, for the translation of tense, aspect, and modality. It was found that the support-vector machine method was the most precise of all the methods tested.
We have developed a new method for Japanese-to-English translation of tense, aspect, and modality that uses an example-based method. In this method the similarity between input and example sentences is defined as the degree of semantic matching between the expressions at the ends of the sentences. Our method also uses the k-nearest neighbor method in order to exclude the effects of noise; for example, wrongly tagged data in the bilingual corpora. Experiments show that our method can translate tenses, aspects, and modalities more accurately than the top-level MT software currently available on …
George A. Miller said that human beings have only seven chunks in short-term memory, plus or minus two. We counted the number of bunsetsus (phrases) whose modifiees are undetermined in each step of an analysis of the dependency structure of Japanese sentences, and which therefore must be stored in short-term memory. The number was roughly less than nine, the upper bound of seven plus or minus two. We also obtained similar results with English sentences under the assumption that human beings recognize a series of words, such as a noun phrase (NP), as a unit. This indicates that if we assume tha…
Cross-language information retrieval (CLIR), where queries and documents are in different languages, has of late become one of the major topics within the information retrieval community. This paper proposes a Japanese/English CLIR system, where we combine a query translation and retrieval modules. We currently target the retrieval of technical documents, and therefore the performance of our system is highly dependent on the quality of the translation of technical terms. However, the technical term translation is still problematic in that technical terms are often compound words, and thus new …
Recent machine translation algorithms mainly rely on parallel corpora. However, since the availability of parallel corpora remains limited, only some resource-rich language pairs can benefit from them. We constructed a parallel corpus for English-Japanese, for which the amount of publicly available parallel corpora is still limited. We constructed the parallel corpus by broadly crawling the web and automatically aligning parallel sentences. Our collected corpus, called JParaCrawl, amassed over 8.7 million sentence pairs. We show how it includes a broader range of domains and how a neural machi…
J-CRe3: A Japanese Conversation Dataset for Real-world Reference Resolution
Kyoto University（京都大学）；RIKEN（理化学研究所）；Tokyo University of Science（东京理科大学）；Nara Institute of Science and Technology（奈良科学技术研究所）；National Institute of Informatics（情报通信研究机构）
Indian Institute of Technology, Ropar（印度理工学院罗帕尔分校）；CSIR-Central Scientific Instruments Organisation（CSIR中央科学仪器组织）；Academy of Scientific and Innovative Research（科学与创新研究院）；Tokai Gakuen University（东海学园大学）；Tokyo Institute of Technology（东京工业大学）
Moral psychology is a domain that deals with moral identity, appraisals and emotions. Previous work has primarily focused on moral development and the associated role of culture. Knowing that language is an inherent element of a culture, we used the social media platform Twitter to compare moral behaviors of Japanese tweets with English tweets. The five basic moral foundations, i.e., Care, Fairness, Ingroup, Authority and Purity, along with the associated emotional valence were compared between English and Japanese tweets. The tweets from Japanese users depicted relatively higher Fairness, Ing…
The purpose of this work is to contribute toward the larger goal of creating a Quantum Natural Language Processing (QNLP) translator program. This work contributes original diagrammatic representations of the Japanese language based on prior work that accomplished on the English language based on category theory. The germane differences between the English and Japanese languages are emphasized to help address English language bias in the current body of research. Additionally, topological principles of these diagrams and many potential avenues for further research are proposed. Why is this end…
