Commercial Machine Translation Research Articles

A linguistic approach to Natural Language Generation: The language spectrum, language models, and OpenLogos

Natural Language Generation, a well-established research area within the field of Natural Language Processing, has achieved significant milestones such as in machine translation and paraphrasing. Large Language Models, while improving text quality, show varying effectiveness across different linguistic registers and cultural contexts, raising concerns mostly due to (i) difficulties in understanding the internal workings of the models; (ii) difficulties in understanding the generative processes, which are not transparent; (iii) difficulties in human intervention in the generative function; (iv) propensity for hallucination and incorrect information content, (v) potential irresponsible use of extensive resources from unidentified sources, and (vi) the risk of misuse, among the most obvious hurdles. This article has three main goals: (1) suggest a linguistic approach to Natural Language Generation, (2) explore the broad language spectrum, including formal to informal styles and objective to subjective language, among others, and (3) propose language models inspired by the Logos Model for enhanced transparency, traceability, and customization. The Logos Model is at the core of the best-documented pioneer commercial machine translation system called Logos, which preludes the history of Generative Artificial Intelligence. The open-source version, OpenLogos, offers valuable resources for machine translation, paraphrasing, abstractive summarization, and various other tasks related to Natural Language Generation, and it serves as an excellent resource for training researchers across multiple fields associated with Artificial Intelligence. The OpenLogos resources were made publicly available through the Multi3Generation COST Action (CA18231).

Good, but not always Fair: An Evaluation of Gender Bias for three Commercial Machine Translation Systems

Machine Translation (MT) continues to make significant strides in quality and is increasingly adopted on a larger scale. Consequently, analyses have been redirected to more nuanced aspects, intricate phenomena, as well as potential risks that may arise from the widespread use of MT tools. Along this line, this paper offers a meticulous assessment of three commercial MT systems - Google Translate, DeepL, and Modern MT - with a specific focus on gender translation and bias. For three language pairs (English-Spanish, English-Italian, and English-French), we scrutinize the behavior of such systems at several levels of granularity and on a variety of naturally occurring gender phenomena in translation. Our study takes stock of the current state of online MT tools, by revealing significant discrepancies in the gender translation of the three systems, with each system displaying varying degrees of bias despite their overall translation quality.

Natural Language Processing and Computational Linguistics

Natural Language Processing and Computational Linguistics

Employing word mover's distance for cross‐lingual plagiarized text detection

AbstractWhile the blossom of globalization has achieved many successes, cross‐lingual plagiarism, however, has become popular. Detecting plagiarism across languages requires the capability of comparing semantical similarities between texts of different languages. Previous works rely on massive bilingual resources such as comparable corpus, parallel corpus, and even commercial machine translation as references. However, towards domain‐specific applications, collecting such resources is labor‐intensive and impractical. Also, the absence of interpretability of existing methods lead to the difficulty of investigating retrieval results by humans. Hence, it is imperative to have a resource‐light and interpretable method for cross‐lingual plagiarism detection. In this study, we propose a new detection method, called CL‐WMD, which is built upon word embedding techniques. CL‐WMD requires only a small set of translation pairs to constitute a bilingual reference and calculates semantical distances between texts by word mover's distance, which can provide explicable word alignment information between two compared text spans. Our experiments are conducted under a bilingual scientific publication corpus composed of two typologically diverse languages: English and Chinese. The results demonstrate that CL‐WMD has higher accuracy than most existing methods and achieves better or comparable performance when compared to the translation‐based method in paragraph‐level and sentence‐level plagiarism detection tasks.

Integration of a Multilingual Preordering Component into a Commercial SMT Platform

AbstractWe present a multilingual preordering component tailored for a commercial Statistical Machine translation platform. In commercial settings, issues such as processing speed as well as the ability to adapt models to the customers’ needs play a significant role and have a big impact on the choice of approaches that are added to the custom pipeline to deal with specific problems such as long-range reorderings.We developed a fast and customisable preordering component, also available as an open-source tool, which comes along with a generic implementation that is restricted neither to the translation platform nor to the Machine Translation paradigm. We test preordering on three language pairs: English →Japanese/German/Chinese for both Statistical Machine Translation (SMT) and Neural Machine Translation (NMT). Our experiments confirm previously reported improvements in the SMT output when the models are trained on preordered data, but they also show that preordering does not improve NMT.

How much hybridization does machine translation Need?

Rule‐based and corpus‐based machine translation (MT) have coexisted for more than 20 years. Recently, boundaries between the two paradigms have narrowed and hybrid approaches are gaining interest from both academia and businesses. However, since hybrid approaches involve the multidisciplinary interaction of linguists, computer scientists, engineers, and information specialists, understandably a number of issues exist.While statistical methods currently dominate research work in MT, most commercial MT systems are technically hybrid systems. The research community should investigate the benefits and questions surrounding the hybridization of MT systems more actively. This paper discusses various issues related to hybrid MT including its origins, architectures, achievements, and frustrations experienced in the community. It can be said that both rule‐based and corpus‐ based MT systems have benefited from hybridization when effectively integrated. In fact, many of the current rule/corpus‐based MT approaches are already hybridized since they do include statistics/rules at some point.

Understanding the economics of machine translation

In recent years, machine translation (MT) has begun to take its place alongside translation memory, terminology management and workflow automation as a key tool in large-scale professional translation projects. Nevertheless, misconceptions abound about the ways in which MT affects the professional translation work process. Effective, efficient and appropriate use of MT in professional translation production is a complex affair that requires an awareness and understanding of the factors that shape the success of MT deployments in terms of both systems development and return on investment (ROI). Generally speaking, however, these factors are not well understood by translation buyers, translation agencies or professional translators. This article provides an overview of the factors that shape successful commercial MT use and addresses some common misperceptions surrounding MT in professional translation projects.

Retrieval effectiveness of machine translated queries

AbstractThis article describes and evaluates various information retrieval models used to search document collections written in English through submitting queries written in various other languages, either members of the Indo‐European family (English, French, German, and Spanish) or radically different language groups such as Chinese. This evaluation method involves searching a rather large number of topics (around 300) and using two commercial machine translation systems to translate across the language barriers. In this study, mean average precision is used to measure variances in retrieval effectiveness when a query language differs from the document language. Although performance differences are rather large for certain languages pairs, this does not mean that bilingual search methods are not commercially viable. Causes of the difficulties incurred when searching or during translation are analyzed and the results of concrete examples are explained.

A Methodology for Evaluating Arabic Machine Translation Systems

This paper presents a methodology for evaluating Arabic Machine Translation (MT) systems. We are specifically interested in evaluating lexical coverage, grammatical coverage, semantic correctness and pronoun resolution correctness. The methodology presented is statistical and is based on earlier work on evaluating MT lexicons in which the idea of the importance of a specific word sense to a given application domain and how its presence or absence in the lexicon affects the MT system's lexical quality, which in turn will affect the overall system output quality. The same idea is used in this paper and generalized so as to apply to grammatical coverage, semantic correctness and correctness of pronoun resolution. The approach adopted in this paper has been implemented and applied to evaluating four English-Arabic commercial MT systems. The results of the evaluation of these systems are presented for the domain of the Internet and Arabization.

Statistical Models for Monolingual and Bilingual Information Retrieval

This work reviews information retrieval systems developed at ITC-irst which were evaluated through several tracks of CLEF, during the last three years. The presentation tries to follow the progress made over time in developing new statistical models first for monolingual information retrieval, then for cross-language information retrieval. Besides describing the underlying theory, performance of monolingual and bilingual information retrieval models are reported, respectively, on Italian monolingual tracks and Italian-English bilingual tracks of CLEF. Monolingual systems by ITC-irst performed consistently well in all the official evaluations, while the bilingual system ranked in CLEF 2002 just behind competitors using commercial machine translation engines. However, by experimentally comparing our statistical topic translation model against a state-of-the-art commercial system, no statistically significant difference in retrieval performance could be measured on a larger set of queries.

Embedding Web-Based Statistical Translation Models in Cross-Language Information Retrieval

Although more and more language pairs are covered by machine translation (MT) services, there are still many pairs that lack translation resources. Cross-language information retrieval (CLIR) is an application that needs translation functionality of a relatively low level of sophistication, since current models for information retrieval (IR) are still based on a bag of words. The Web provides a vast resource for the automatic construction of parallel corpora that can be used to train statistical translation models automatically. The resulting translation models can be embedded in several ways in a retrieval model. In this article, we will investigate the problem of automatically mining parallel texts from the Web and different ways of integrating the translation models within the retrieval process. Our experiments on standard test collections for CLIR show that the Web-based translation models can surpass commercial MT systems in CLIR tasks. These results open the perspective of constructing a fully automatic query translation device for CLIR at a very low cost.

An automatic English-Arabic HTML page translation system

The Internet and the World Wide Web have become an integral part of everyday life, an important source of information and a communication medium. One of the main problems confronting non-English speakers in using the Internet is that it is heavily dominated by the English language. Knowledge of English is a prerequisite for using the Web efficiently and benefiting from the multitude of services it offers. This paper introduces the approach underlying a prototype system that translates Web pages from English to Arabic automatically. The system uses a commercial machine translation system to translate the textual part of a Web page. It then displays a Web page containing the Arabic translation with all tags inserted in the right places so that the layout and content of the original (English) page are preserved.

Analysis of lengthy sentences using an english comparative structure model

AbstractToday, technological expertise gained from studies on natural languages is applied to various practical systems, including machine translation systems. English‐Japanese machine translation, however, still remains difficult in the translation of lengthy sentences that contain multiple complicated expressions. Due to limited information volume installed in current machine translation systems, such systems cannot handle the complicated sentences actually used in English. This paper discusses comparative structures, which frequently appear in long English sentences, clarifies characteristics of such structures through grammatical studies and the statistical analysis of the sentences contained in Corpus, and explains the development of comparative structure models and a new analysis method using such models. Having more flexibility than the conventional machine translation method, the new method proposed in this paper has been proved suitable for long sentences for which the grammatical rules and vocabulary installed in conventional machine translation systems are insufficient. English newspaper articles were analyzed with a commercial machine translation system and the new method. The result showed that the new method provided translations that were 41.6 percent more accurate than the conventional system translations.