No file available
To read the file of this research,
you can request a copy directly from the author.
Artificial intelligence and translation industry
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
This article deals with AI and translation industry ............
ResearchGate has not been able to resolve any citations for this publication.
Machine translation (MT) is the automated process of translating text between different languages, encompassing a wide range of language pairs. This study focuses on non-professional bilingual speakers of Turkish and English, aiming to assess their ability to discern accuracy in machine translations and their preferences regarding MT. A particular emphasis is placed on the linguistically subtle yet semantically meaningful concept of evidentiality. In this experimental investigation, 36 Turkish–English bilinguals, comprising both early and late bilinguals, were presented with simple declarative sentences. These sentences varied in their evidential meaning, distinguishing between firsthand and non-firsthand evidence. The participants were then provided with MT of these sentences in both translation directions (Turkish to English and English to Turkish) and asked to identify the accuracy of these translations. Additionally, participants were queried about their preference for MT in four crucial domains: medical, legal, academic, and daily contexts. The findings of this study indicated that late bilinguals exhibited a superior ability to detect translation accuracy, particularly in the case of firsthand evidence translations, compared to their early bilingual counterparts. Concerning the preference for MT, age of acquisition and the accuracy detection of non-firsthand sentence translations emerged as significant predictors.
In this comprehensive study, a novel MATLAB to Python (M-to-PY) conversion process is showcased, specifically tailored for an intricate image skeletonization project involving fifteen MATLAB files and a large dataset. The central innovation of this research is the adept use of ChatGPT-4 as an AI assistant, pivotal in crafting a prototype M-to-PY converter. This converter’s capabilities were thoroughly evaluated using a set of test cases generated by the Bard bot, ensuring a robust and effective tool. The culmination of this effort was the development of the Skeleton App, adept at image sketching and skeletonization. This live and publicly available app underscores the enormous potential of AI in enhancing the transition of scientific research from MATLAB to Python. The study highlights the blend of AI’s computational prowess and human ingenuity in computational research, making significant strides in AI-assisted scientific exploration and tool development.
As artificial intelligence (AI) translation technology advances, big data, cloud computing, and emerging technologies have enhanced the progress of the data industry over the past several decades. Human-machine translation becomes a new interactive mode between humans and machines and plays an essential role in transmitting information. Nevertheless, several translation models have their drawbacks and limitations, such as error rates and inaccuracy, and they are not able to adapt to the various demands of different groups. Taking the AI-based translation model as the research object, this study conducted an analysis of attention mechanisms and relevant technical means, examined the setbacks of conventional translation models, and proposed an AI-based translation model that produced a clear and high quality translation and presented a reference to further perfect AI-based translation models. The values of the manual and automated evaluation have demonstrated that the human-machine translation model improved the mismatchings between texts and contexts and enhanced the accurate and efficient intelligent recognition and expressions. It is set to a score of 1-10 for evaluation comparison with 30 language users as participants, and the achieved 6 points or above is considered effective. The research results suggested that the language fluency score rose from 4.9667 for conventional Statistical Machine Translation to 6.6333 for the AI-based translation model. As a result, the human-machine translation model improved the efficiency, speed, precision, and accuracy of language input to a certain degree, strengthened the correlation between semantic characteristics and intelligent recognition, and pushed the advancement of intelligent recognition. It can provide accurate and high-quality translation for language users and achieve an understanding of natural language input and output and automatic processing.
The rapid global expansion of ChatGPT, which plays a crucial role in interactive knowledge sharing and translation, underscores the importance of comparative performance assessments in artificial intelligence (AI) technology. This study concentrated on this crucial issue by exploring and contrasting the translation performances of large language models (LLMs) and neural machine translation (NMT) systems. For this aim, the APIs of Google Translate, Microsoft Translator, and OpenAI’s ChatGPT were utilized, leveraging parallel corpora from the Workshop on Machine Translation (WMT) 2018 and 2020 benchmarks. By applying recognized evaluation metrics such as BLEU, chrF, and TER, a comprehensive performance analysis across a variety of language pairs, translation directions, and reference token sizes was conducted. The findings reveal that while Google Translate and Microsoft Translator generally surpass ChatGPT in terms of their BLEU, chrF, and TER scores, ChatGPT exhibits superior performance in specific language pairs. Translations from non-English to English consistently yielded better results across all three systems compared with translations from English to non-English. Significantly, an improvement in translation system performance was observed as the token size increased, hinting at the potential benefits of training models on larger token sizes.
This article investigates the impact of ChatGPT on the current language regimes in Europe for patent translation. With the growing number of patent applications being filed annually, there is a high demand for cost-effective and high-quality patent translation services. Despite advancements in translation technology, including ChatGPT, there remains a scarcity of research on the use of these technologies in patent translation, particularly with regard to the legal validity of machine-generated translations. The article takes an interdisciplinary approach, incorporating a review of regulatory documents, a preliminary evaluation of ChatGPT’s performance using the LISA QA matrix and a review of relevant doctrinal literature. The regulatory examination in this study centers on the provisions of the European Patent Convention and related legal documents that specify the language criteria for acquiring a patent within Europe. The preliminary evaluation uses a sample of 20 patents translated from English into Portuguese to assess accuracy, fluency and acceptability. The examination of available legal literature strives to attain a thorough comprehension of how laws and principles of the legal system are being put into practice in association with the progression of translation technologies. This article endeavors to examine the potential effects of utilizing ChatGPT in patent translation on existing language regulations in Europe and if new regulations will be necessary. It highlights the need for further research in this field to better understand the advantages and implications of using advanced language models in patent translation.
The current paper reports on the results of a national research project investigating the use of didactic audiovisual translation (DAT) in foreign language teaching. Although previous research in this field concludes that there are positive outcomes in students' learning, most studies rely on small samples and analyse one language skill only. The TRADILEX project aims at examining the effect of several modes of audiovisual translation (AVT)-i.e., subtitling, dubbing, subtitling for the deaf and hard of hearing, audio description and voiceover-in oral and written skills of university students learning English as a foreign language. This article assesses the effectiveness of a didactic proposal that includes 30 lesson plans through an intervention carried out with 566 participants from eight universities in Spain. The study relies on a quantitative design, and statistical tests (descriptive statistics and Wilcoxon Test) were carried out to estimate the effect of DAT on oral and written reception and oral and written production of the students. Our results underline that there are statistically significant differences that confirm students improved in the four communicative skills in the foreign language. Besides, there is a positive evolution in students' achievement during the study, and participants report a favourable perception of the didactic intervention. Resumen Este trabajo presenta los resultados de un proyecto nacional sobre el uso de la traducción audiovisual didáctica (TAD) en el aprendizaje de lenguas extranjeras. Si bien los resultados de investigaciones previas en este campo son favorables con respecto al desempeño académico del alumnado, la mayoría de los estudios analizan una única destreza lingüística y cuentan con muestras reducidas. El proyecto TRADILEX tiene como objetivo analizar el efecto del uso de diferentes modalidades de traducción audiovisual (TAV) (subtitulado, doblaje, subtitulado para sordos, audiodescripción y voces superpuestas) en las competencias orales y escritas del alumnado universitario que estudia inglés como lengua extranjera. El presente artículo analiza la efectividad de una propuesta didáctica que incluye 30 planes de estudio y que se analizó a través de una intervención con 566 participantes de ocho universidades españolas. La investigación toma un diseño cuantitativo en el que se realizaron pruebas estadísticas (estadísticos descriptivos y Test de Wilcoxon) para valorar el efecto de la TAD sobre las destrezas lingüísticas de recepción oral y escrita, producción oral y escrita del alumnado. Los resultados subrayan que hay diferencias estadísticamente significativas que demuestran que los participantes del estudio mejoran en las cuatro destrezas comunicativas analizadas. Además, los datos también confirman la evaluación positiva del rendimiento del alumnado en el transcurso del estudio, así como una percepción favorable de la intervención didáctica por parte de los participantes.
ChatGPT, an AI chatbot developed by OpenAI and powered by advanced large language models, has attracted significant interest across a range of industries. With the more recent release of GPT-4, the field of language understanding and generation continues to progress rapidly. In this article, we delve into the use cases, strengths, and limitations of both ChatGPT and GPT-4, assessing their potential impact within the context of the translation industry. We emphasize the potential benefits these platforms offer to language professionals, while underlining the ongoing need for human expertise. Furthermore, we provide recommendations for practitioners to adapt to the emergence of these tools and suggest avenues for future development. Our work aims to foster a comprehensive understanding of the role of ChatGPT and GPT-4 within the translation sector, and encourage informed discussions about their potential and implications.
Automatic translation from signed to spoken languages is an interdisciplinary research domain on the intersection of computer vision, machine translation (MT), and linguistics. While the domain is growing in terms of popularity—the majority of scientific papers on sign language (SL) translation have been published in the past five years—research in this domain is performed mostly by computer scientists in isolation. This article presents an extensive and cross-domain overview of the work on SL translation. We first give a high level introduction to SL linguistics and MT to illustrate the requirements of automatic SL translation. Then, we present a systematic literature review of the state of the art in the domain. Finally, we outline important challenges for future research. We find that significant advances have been made on the shoulders of spoken language MT research. However, current approaches often lack linguistic motivation or are not adapted to the different characteristics of SLs. We explore challenges related to the representation of SL data, the collection of datasets and the evaluation of SL translation models. We advocate for interdisciplinary research and for grounding future research in linguistic analysis of SLs. Furthermore, the inclusion of deaf and hearing end users of SL translation applications in use case identification, data collection, and evaluation, is of utmost importance in the creation of useful SL translation models.
We evaluated the performance of a large language model called ChatGPT on the United States Medical Licensing Exam (USMLE), which consists of three exams: Step 1, Step 2CK, and Step 3. ChatGPT performed at or near the passing threshold for all three exams without any specialized training or reinforcement. Additionally, ChatGPT demonstrated a high level of concordance and insight in its explanations. These results suggest that large language models may have the potential to assist with medical education, and potentially, clinical decision-making.
Languages that allow free word order, such as Arabic dialects, are of significant difficulty for neural machine translation (NMT) because of many scarce words and the inefficiency of NMT systems to translate these words. Unknown Word (UNK) tokens represent the out-of-vocabulary words for the reason that NMT systems run with vocabulary that has fixed size. Scarce words are encoded completely as sequences of subword pieces employing the Word-Piece Model. This research paper introduces the first Transformer-based neural machine translation model for Arabic vernaculars that employs subword units. The proposed solution is based on the Transformer model that has been presented lately. The use of subword units and shared vocabulary within the Arabic dialect (the source language) and modern standard Arabic (the target language) enhances the behavior of the multi-head attention sublayers for the encoder by obtaining the overall dependencies between words of input sentence for Arabic vernacular. Experiments are carried out from Levantine Arabic vernacular (LEV) to modern standard Arabic (MSA) and Maghrebi Arabic vernacular (MAG) to MSA, Gulf–MSA, Nile–MSA, Iraqi Arabic (IRQ) to MSA translation tasks. Extensive experiments confirm that the suggested model adequately addresses the unknown word issue and boosts the quality of translation from Arabic vernaculars to Modern standard Arabic (MSA).
A human-inspired, linguistically sophisticated model of language understanding for intelligent agent systems.
One of the original goals of artificial intelligence research was to endow intelligent agents with human-level natural language capabilities. Recent AI research, however, has focused on applying statistical and machine learning approaches to big data rather than attempting to model what people do and how they do it. In this book, Marjorie McShane and Sergei Nirenburg return to the original goal of recreating human-level intelligence in a machine. They present a human-inspired, linguistically sophisticated model of language understanding for intelligent agent systems that emphasizes meaning—the deep, context-sensitive meaning that a person derives from spoken or written language.
With Linguistics for the Age of AI, McShane and Nirenburg offer a roadmap for creating language-endowed intelligent agents (LEIAs) that can understand,explain, and learn. They describe the language-understanding capabilities of LEIAs from the perspectives of cognitive modeling and system building, emphasizing “actionability”—which involves achieving interpretations that are sufficiently deep, precise, and confident to support reasoning about action. After detailing their microtheories for topics such as semantic analysis, basic coreference, and situational reasoning, McShane and Nirenburg turn to agent applications developed using those microtheories and evaluations of a LEIA's language understanding capabilities.
McShane and Nirenburg argue that the only way to achieve human-level language understanding by machines is to place linguistics front and center, using statistics and big data as contributing resources. They lay out a long-term research program that addresses linguistics and real-world reasoning together, within a comprehensive cognitive architecture.
The open access edition of this book was made possible by generous funding from Arcadia – a charitable fund of Lisbet Rausing and Peter Baldwin.
This study strives to answer one major question: Does culture translate?, employing the translation of English proverbs into Arabic by senior students majoring in English. The study involves 30 English proverbs collected from different sources, based on three criteria, namely complete Arabic equivalence, partial Arabic equivalence and zero Arabic equivalence. These 30 proverbs were distributed to 20 randomly selected senior students as participants. The participants translated the 30 proverbs in the form of a translation test in two phases. The results of the study show that teaching the cultural aspects of the proverbs has developed and improved the participants’ translations considerably, thus providing empirical evidence that culture translates .
Fluorescein angiography (FA) is a procedure used to image the vascular structure of the retina and requires the insertion of an exogenous dye with potential adverse side effects. Currently, there is only one alternative non-invasive system based on Optical coherence tomography (OCT) technology, called OCT angiography (OCTA), capable of visualizing retina vasculature. However, due to its cost and limited view, OCTA technology is not widely used. Retinal fundus photography is a safe imaging technique used for capturing the overall structure of the retina. In order to visualize retinal vasculature without the need for FA and in a cost-effective, non-invasive, and accurate manner, we propose a deep learning conditional generative adversarial network (GAN) capable of producing FA images from fundus photographs. The proposed GAN produces anatomically accurate angiograms, with similar fidelity to FA images, and significantly outperforms two other state-of-the-art generative algorithms ( p < . 001 and p < . 0001 ). Furthermore, evaluations by experts shows that our proposed model produces such high quality FA images that are indistinguishable from real angiograms. Our model as the first application of artificial intelligence and deep learning to medical image translation, by employing a theoretical framework capable of establishing a shared feature-space between two domains (i.e. funduscopy and fluorescein angiography) provides an unrivaled way for the translation of images from one domain to the other.
Despite the recent success of deep neural networks in natural language processing and other spheres of artificial intelligence, their interpretability remains a challenge. We analyze the representations learned by neural machine translation (NMT) models at various levels of granularity and evaluate their quality through relevant extrinsic properties. In particular, we seek answers to the following questions: (i) How accurately is word structure captured within the learned representations, which is an important aspect in translating morphologically rich languages? (ii) Do the representations capture long-range dependencies, and effectively handle syntactically divergent languages? (iii) Do the representations capture lexical semantics? We conduct a thorough investigation along several parameters: (i) Which layers in the architecture capture each of these linguistic phenomena; (ii) How does the choice of translation unit (word, character, or subword unit) impact the linguistic properties captured by the underlying representations? (iii) Do the encoder and decoder learn differently and independently? (iv) Do the representations learned by multilingual NMT models capture the same amount of linguistic information as their bilingual counterparts? Our data-driven, quantitative evaluation illuminates important aspects in NMT models and their ability to capture various linguistic phenomena. We show that deep NMT models trained in an end-to-end fashion, without being provided any direct supervision during the training process, learn a non-trivial amount of linguistic information. Notable findings include the following observations: (i) Word morphology and part-of-speech information are captured at the lower layers of the model; (ii) In contrast, lexical semantics or non-local syntactic and semantic dependencies are better represented at the higher layers of the model; (iii) Representations learned using characters are more informed about word-morphology compared to those learned using subword units; and (iv) Representations learned by multilingual models are richer compared to bilingual models.
The use of neural machine translation (NMT) in a professional scenario implies a number of challenges despite growing evidence that, in language combinations such as English to Spanish, NMT output quality has already outperformed statistical machine translation in terms of automatic metric scores. This article presents the result of an empirical test that aims to shed light on the differences between NMT post-editing and translation with the aid of a translation memory (TM). The results show that NMT post-editing involves less editing than TM segments, but this editing appears to take more time, with the consequence that NMT post-editing does not seem to improve productivity as may have been expected. This might be due to the fact that NMT segments show a higher variability in terms of quality and time invested in post-editing than TM segments that are ‘more similar’ on average. Finally, results show that translators who perceive that NMT boosts their productivity actually performed faster than those who perceive that NMT slows them down.
It is intuitive that semantic representations can be useful for machine translation, mainly because they can help in enforcing meaning preservation and handling data sparsity (many sentences correspond to one meaning) of machine translation models. On the other hand, little work has been done on leveraging semantics for neural machine translation (NMT). In this work, we study the usefulness of AMR (abstract meaning representation) on NMT. Experiments on a standard English-to-German dataset show that incorporating AMR as additional knowledge can significantly improve a strong attention-based sequence-to-sequence neural translation model.
The issue of how a null subject (or argumental pro) is licensed and interpreted has been a matter of debate in syntactic theory for decades. Participating in this debate, this paper proposes a novel approach to the licensing and interpretation of the referential 3 (person) pro/topic in human languages, based on discourse-syntax interface. I provide evidence from across languages that the antecedent of the dropped pro/topic is a (silent) preverbal DP. I show that this DP is an aboutness-topic, merged in the C-domain, specifically in the Specifier of the Topic Phrase, where Topic Phrase is assumed to be a phase whose head, i.e. Topic, is endowed with an aboutness feature. Based on interpretation motivations, the aboutness feature counts as an Edge Feature, which requires merging a (silent) aboutness-topic in Spec,TopP, hence yielding a discourse property and coreferentially correlating the given aboutness-topic with the argumental pro in Spec,vP. Evidence is provided that pro enters the derivation with valued, but uninterpretable features. These valued features (of pro) value T’s unvalued features via agree. The uninterpretable features of pro are interpreted by the interpretable features of the A-topic via agree as Match. As a result, pro is interpreted as a definite 3 person pronoun. The paper also provides empirical evidence that pro can be locally and nonlocally coreferentially correlated with the A-topic via matching A`-chains. A (Silent) A-topic Principle is proposed as a Universal Grammar condition, which is necessitated by interpretive and performative requirements. Given this Universal Grammar property, the Silent A-topic Principle licenses (silent) A-topics as antecedents for pros/dropped topics across human languages.
In this article we present a three-step methodology for dynamically improving a statistical machine translation (SMT) system by incorporating human feedback in the form of free edits on the system translations. We target at feedback provided by casual users, which is typically error-prone. Thus, we first propose a filtering step to automatically identify the better user-edited translations and discard the useless ones. A second step produces a pivot-based alignment between source and user-edited sentences, focusing on the errors made by the system. Finally, a third step produces a new translation model and combines it linearly with the one from the original system. We perform a thorough evaluation on a real-world dataset collected from the Reverso.net translation service and show that every step in our methodology contributes significantly to improve a general purpose SMT system. Interestingly, the quality improvement is not only due to the increase of lexical coverage, but to a better lexical selection, reordering, and morphology. Finally, we show the robustness of the methodology by applying it to a different scenario, in which the new examples come from an automatically Web-crawled parallel corpus. Using exactly the same architecture and models provides again a significant improvement of the translation quality of a general purpose baseline SMT system.
Scientometrics is the study of the quantitative aspects of the process of
science as a communication system. It is centrally, but not only, concerned
with the analysis of citations in the academic literature. In recent years it
has come to play a major role in the measurement and evaluation of research
performance. In this review we consider: the historical development of
scientometrics, sources of citation data, citation metrics and the "laws" of
scientometrics, normalisation, journal impact factors and other journal
metrics, visualising and mapping science, evaluation and policy, and future
developments.
We describe a series of five statistical models of the translation process and give algorithms for estimating the parameters of these models given a set of pairs of sentences that are translations of one another. We define a concept of word-by-word alignment between such pairs of sentences. For any given pair of such sentences each of our models assigns a probability to each of the possible word-by-word alignments. We give an algorithm for seeking the most probable of these alignments. Although the algorithm is suboptimal, the alignment thus obtained accounts well for the word-by-word relationships in the pair of sentences. We have a great deal of data in French and English from the proceedings of the Canadian Parliament. Accordingly, we have restricted our work to these two languages; but we feel that because our algorithms have minimal linguistic content they would work well on other pairs of languages. We also feel, again because of the minimal linguistic content of our algorithms, that it is reasonable to argue that word-by-word alignments are inherent in any sufficiently large bilingual corpus.
Apertium is a free/open-source platform for rule-based machine translation. It is being widely used to build machine translation
systems for a variety of language pairs, especially in those cases (mainly with related-language pairs) where shallow transfer
suffices to produce good quality translations, although it has also proven useful in assimilation scenarios with more distant
pairs involved. This article summarises the Apertium platform: the translation engine, the encoding of linguistic data, and
the tools developed around the platform. The present limitations of the platform and the challenges posed for the coming years
are also discussed. Finally, evaluation results for some of the most active language pairs are presented. An appendix describes
Apertium as a free/open-source project.
KeywordsFree/open-source machine translation–Rule-based machine translation–Apertium–Shallow transfer–Finite-state transducers
A phrase-based statistical machine translation approach — the alignment template approach — is described. This translation approach allows for general many-to-many relations between words. Thereby, the context of words is taken into account in the translation model, and local changes in word order from source to target language can be learned explicitly. The model is described using a log-linear modeling approach, which is a generalization of the often used source-channel approach. Thereby, the model is easier to extend than classical statistical machine translation systems. We describe in detail the process for learning phrasal translations, the feature functions used, and the search algorithm. The evaluation of this approach is performed on three different tasks. For the German-English speech Verbmobil task, we analyze the effect of various system components. On the French-English Canadian Hansards task, the alignment template system obtains significantly better results than a single-word-based translation model. In the Chinese-English 2002 National Institute of Standards and Technology (NIST) machine translation evaluation it yields statistically significantly better NIST scores than all competing research and commercial translation systems.
This article describes the latest development of a generic approach to detecting and visualizing emerging trends and transient patterns in scientific literature. The work makes substantial theoretical and methodological contributions to progressive knowledge domain visualization. A specialty is conceptualized and visualized as a time-variant duality between two fundamental concepts in information science – research fronts and intellectual bases. A research front is defined as an emergent and transient grouping of concepts and underlying research issues. The intellectual base of a research front is its citation and co-citation footprint in scientific literature – an evolving network of scientific publications cited by research front concepts. Kleinberg’s burst detection algorithm is adapted to identify emergent research front concepts. Freeman’s betweenness centrality metric is used to highlight potential pivotal points of paradigm shift over time. Two complementary visualization views are designed and implemented: cluster views and time-zone views. The contributions of the approach are: 1) the nature of an intellectual base is algorithmically and temporally identified by emergent research-front terms, 2) the value of a co-citation cluster is explicitly interpreted in terms of research front concepts and 3) visually prominent and algorithmically detected pivotal points substantially reduce the complexity of a visualized network. The modeling and visualization process is implemented in CiteSpace II, a Java application, and applied to the analysis of two research fields: mass extinction (1981-2004) and terrorism (1990-2003). Prominent trends and pivotal points in visualized networks were verified in collaboration with domain experts, who are the authors of pivotal-point articles. Practical implications of the work are discussed. A number of challenges and opportunities for future studies are identified.
Dimensions ID: pub.1001131784
this paper, we present a statistical approach to machine translation. We describe the application of our approach to translation from French to English and give preliminary results
This study sets out to answer one major question: Can ChatGPT capture swearing nuances? It presents an empirical study on the ability of ChatGPT to translate Arabic oath expressions into English. 30 Arabic oath expressions were collected from the literature. These 30 oaths were first translated via ChatGPT and then analyzed and compared to the human translation in terms of types of gaps left unfulfilled by ChatGPT. Specifically, the gaps involved are: religious gap, cultural gap, both religious and cultural gaps, no gap, using non-oath particles, redundancy and noncapturing of Arabic script diacritics. It concludes that ChatGPT translation of oaths is still much unsatisfactory, unveiling the need of further developments of ChatGPT, and the inclusion of Arabic data on which ChatGPT should be trained including oath expressions, oath nuances, rituals, and practices.
The study of financial stability is a highly expansive and significant area of academic inquiry. However, traditional literature reviews are often constrained in their scope as they tend to provide fragmented insights from a subset of the overall corpus of financial stability. In order to bridge this existing knowledge gap, the objective of this study is to undertake a comprehensive review on a large scale, with the aim of offering an up-to-date analysis of the performance and intellectual framework concerning the broad concept of financial stability. The present study undertakes a comprehensive examination of the existing body of literature on financial stability, employing a scientometric survey methodology to analyze scholarly research in this field. The study proposes a research agenda based on two prominent developments identified in the field: systemic risk and macroprudential policy. These findings highlight the importance of conducting additional studies on how various factors affect the sustainability of financial security and risk, such as the financial hazards associated with climate change. There is also an emphasis on macroprudential policy and its possible role in preserving financial security. To maintain macroeconomic stability, we recommend that macroprudential policy should be employed to deal with systemic risk and the credit/financial cycle. We recognize the need for further research into the proper calibration of macroprudential instruments and the limited empirical evidence on the effectiveness of macroprudential regulation. To fully grasp the dynamics at play between climate change, financial stability, and macroprudential policy, the article concludes with a need for a multidisciplinary approach.
Although automated translation has been available for decades in myriad forms, the implication of the current exponential advancement in artificial intelligence (AI) for communication in general and translation in particular is more starkly affrontational than ever. Although Large Language Models, of which ChatGPT is exemplary, were not specifically designed for translation purposes, they are attested to have attained a sufficient degree of technical sophistication as to generate translations that match or surpass dedicated translation systems in the market like Google Translate and DeepL. This impacts the modus operandi of communication and the self-concept of language professionals including, of course, translators. This article asks how translation as a field of practice can best respond to this development. It critically reflects on the implications of AI for the conception of translation, arguing that an alternative framing around the idea of distribution allows us to rescale translation toward broader competencies and conceive of AI as a prosthesis of translators’ minds. The article advocates a posthumanist perspective on translation with a view to expanding its spectrum of skills, modes, and media as well as transcending the traditional personae of translators.
This study examines the performance of the neural machine translation system DeepL in translating Shakespeare’s plays Coriolanus and The Merchant of Venice . The aim here is to explore the strengths and limitations of an AI-based English-Chinese translation of literary texts. Adopting a corpus-based approach, the study investigates the accuracy and fluency rates, the linguistic features, and the use of various methods of translation in the Chinese translations of Shakespeare’s plays conducted via DeepL. It compares these to the translations by Liang Shiqiu, a well-known Chinese translator. The study finds that DeepL performs well in translating these works, with an accuracy and fluency rate of above 80% in sampled texts, showing the potential of the use of neural machine translation in translating literary texts across distant languages. Our research further reveals that the DeepL translations exhibit a certain degree of creativity in their use of translation methods such as addition, explicitation, conversion and shift of perspective, and in the use of Chinese sentence-final modal particles, as well as Chinese modal verbs. On the other hand, the system appears to be limited in that a certain amount of translation errors are present, including literal translations.
Neural Machine Translation (NMT) has gained increasing popularity among EFL learners as a CALL tool to improve vocabulary, and many learners have reported its helpfulness for vocabulary learning. However, while there has been some evidence suggesting NMT’s facilitative role in improving learners’ writing on the lexical level, no study has examined whether vocabulary improvement made with the aid of NMT leads to any vocabulary retention when the tool is no longer in use. The present study employed a quasi-experimental design to examine the extent to which editing with NMT may lead to vocabulary improvement, immediate vocabulary retention and delayed vocabulary retention in the legal context and how the findings may vary by learners’ proficiency level. ANOVA results revealed that learners with higher proficiency achieved a similar level of immediate and delayed retention compared to their vocabulary improvement, while the benefits of NMT for learners with lower proficiency were significantly confined to vocabulary improvement and immediate vocabulary retention.
Conversational AI is a game-changer for science. Here’s how to respond. Conversational AI is a game-changer for science. Here’s how to respond.
After more than 70 years of evolution, great achievements have been made in machine translation. Especially in recent years, translation quality has been greatly improved with the emergence of neural machine translation (NMT). In this article, we first review the history of machine translation from rule-based machine translation to example-based machine translation and statistical machine translation. We then introduce NMT in more detail, including the basic framework and the current dominant framework, Transformer, as well as multilingual translation models to deal with the data sparseness problem. In addition, we introduce cutting-edge simultaneous translation methods that achieve a balance between translation quality and latency. We then describe various products and applications of machine translation. At the end of this article, we briefly discuss challenges and future research directions in this field.
Modern deep neural networks achieve impressive performance in engineering applications that require extensive linguistic skills, such as machine translation. This success has sparked interest in probing whether these models are inducing human-like grammatical knowledge from the raw data they are exposed to and, consequently, whether they can shed new light on long-standing debates concerning the innate structure necessary for language acquisition. In this article, we survey representative studies of the syntactic abilities of deep networks and discuss the broader implications that this work has for theoretical linguistics.
Expected final online publication date for the Annual Review of Linguistics, Volume 7 is January 14, 2021. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
Neural Machine Translation (NMT) has gained more and more attention in recent years, mainly due to its simplicity yet state-of-the-art performance. However, previous research has shown that NMT suffers from several limitations: source coverage guidance, translation of rare words, and the limited vocabulary, while Statistical Machine Translation (SMT) has complementary properties that correspond well to these limitations. It is straightforward to improve translation performance by combining the advantages of two kinds of models. This article proposes a general framework for incorporating the SMT word knowledge into NMT to alleviate above word-level limitations. In our framework, the NMT decoder makes more accurate word prediction by referring to the SMT word recommendations in both training and testing phases. Specifically, the SMT model offers informative word recommendations based on the NMT decoding information. Then we use the SMT word predictions as prior knowledge to adjust the NMT word generation probability, which unitizes a neural network-based classifier to digest the discrete word knowledge. In this work, we use two model variants to implement the framework, one with a gating mechanism and the other with a direct competition mechanism. Experimental results on Chinese-to-English and English-to-German translation tasks show that the proposed framework can take advantage of the SMT word knowledge and consistently achieve significant improvements over NMT and SMT baseline systems. IEEE
The hesitant fuzzy linguistic term set (HFLTS) has gained great success as it can be used to represent several linguistic terms or comparative linguistic expressions together with some context-free grammars. This new approach has enabled the analysis and computing of linguistic expressions with uncertainties and opened the door for the possibility to develop more comprehensive and powerful decision theories and methods based on linguistic knowledge. Lots of new approaches and proposals for decision-making problems have been proposed to overcome the limitations of previous linguistic decision-making approaches. Now and in the future, decision-making methodologies and algorithms with hesitant fuzzy linguistic models would be a quite promising research line representing a high-quality breakthrough in this topic. To facilitate the study on HFLTS theory, this paper makes a state-of-the-art survey on HFLTSs based on the 134 selected papers from Web of Sciences published from January 2012 to October 2017. We justify the motivation, definitions, operations, comparison methods and measures of HFLTSs. We also summarize the different extensions of HFLTSs. The studies on multiple criteria decision making (MCDM) with HFLTSs in terms of aggregation operators and MCDM methods are clearly reviewed. We also conduce some overviews on decision making with hesitant fuzzy linguistic preference relations. The applications, research challenges and future directions are also given.
AbstractThe paper describes the process by which the word alignment task performed within SOMAgent works in collaboration with the statistical machine translation system in order to learn a phrase translation table. We studied improvements in the quality of translation using syntax augmented machine translation. We also experimented with different degrees of linguistic analysis from the lexical level to a syntactic or semantic level, in order to generate a more precise alignment. We developed a contextual environment using the Self-Organizing Map, which can model a semantic agent (SOMAgent) that learns the correct meaning of a word used in context in order to deal with specific phenomena such as ambiguity, and to generate more precise alignments that can improve the first choice of the statistical machine translation system giving linguistic knowledge.
A longstanding difficulty for connectionist modeling has been how to represent variable-sized recursive data structures, such as trees and lists, in fixed-width patterns. This paper presents a connectionist architecture which automatically develops compact distributed representations for such compositional structures, as well as efficient accessing mechanisms for them. Patterns which stand for the internal nodes of fixed-valence trees are devised through the recursive use of backpropagation on three-layer auto-associative encoder networks. The resulting representations are novel, in that they combine apparently immiscible aspects of features, pointers, and symbol structures. They form a bridge between the data structures necessary for high-level cognitive tasks and the associative, pattern recognition machinery provided by neural networks.
We describe methods for improving the performance of statistical machine translation (SMT) between four linguistically different languages, i.e., Chinese, English, Japanese, and Korean by using morphosyntactic knowledge. For the purpose of reducing the translation ambiguities and generating grammatically correct and fluent translation output, we address the use of shallow linguistic knowledge, that is: (1) enriching a word with its morphosyntactic features, (2) obtaining shallow linguistically-motivated phrase pairs, (3) iteratively refining word alignment using filtered phrase pairs, and (4) building a language model from morphosyntactically enriched words. Previous studies reported that the introduction of syntactic features into SMT models resulted in only a slight improvement in performance in spite of the heavy computational expense, however, this study demonstrates the effectiveness of morphosyntactic features, when reliable, discriminative features are used. Our experimental results show that word representations that incorporate morphosyntactic features significantly improve the performance of the translation model and language model. Moreover, we show that refining the word alignment using fine-grained phrase pairs is effective in improving system performance.
In this paper, we develop an approach called syntax-based reordering (SBR) to handling the fundamental problem of word ordering for statistical machine translation (SMT). We propose to alleviate the word order challenge including morpho-syntactical and statistical information in the context of a pre-translation reordering framework aimed at capturing short- and long-distance word distortion dependencies. We examine the proposed approach from the theoretical and experimental points of view discussing and analyzing its advantages and limitations in comparison with some of the state-of-the-art reordering methods. In the final part of the paper, we describe the results of applying the syntax-based model to translation tasks with a great need for reordering (Chinese-to-English and Arabic-to-English). The experiments are carried out on standard phrase-based and alternative N-gram-based SMT systems. We first investigate sparse training data scenarios, in which the translation and reordering models are trained on a sparse bilingual data, then scaling the method to a large training set and demonstrating that the improvement in terms of translation quality is maintained.
The following three problems are known to exist with statistical machine translation. (1) the modeling problem involved in prescribing translation relations, (2) the problem of determining parameter settings from a text corpus of translations, and (3) the search problem involved in determining the output text (the translation) given a statistical model and an input text. In this paper we find alignments of translations using phrase-based units in a hierarchical fashion with the intention of solving the above-mentioned modeling and training problems with such hierarchical phrase alignments. As an initial method we perform chunking on the corpus on the basis of these hierarchical alignments, and create translation models using these chunks as translation units. Then, as a second method we convert the translation relations expressed in the hierarchical phrase alignments into correspondences in the translation model, and perform additional training having initialized the model parameters to values obtained from these relations. The results of experiments with Japanese-to-English translation show that both methods improve performance with the second method being particularly effective resulting in an increase in translation rate from 61.3% to 70.0%. © 2007 Wiley Periodicals, Inc. Syst Comp Jpn, 38(6): 70–79, 2007; Published online in Wiley InterScience (www.interscience. wiley.com). DOI 10.1002/scj.20271
We propose a new phrase-based translation model and decoding algorithm that enables us to evaluate and compare several, previously proposed phrase-based translation models.
Improvement in machine translation with generative adversarial networks
- J Ahn
- H Madhu
- V Nguyen
Ahn, J., Madhu, H., & Nguyen, V. (2021). Improvement in machine translation with
generative adversarial networks. arXiv preprint arXiv:2111.15166.
- V Goyle
- P Krishnaswamy
- K G Ravikumar
- U Chattopadhyay
- K Goyle
Goyle, V., Krishnaswamy, P., Ravikumar, K. G., Chattopadhyay, U., & Goyle, K.
(2023). Neural machine translation for low resource languages.arXiv preprint
arXiv:2304.07869.