Post-editing Machine Translation Research Articles

Enhancing Post-Editing Machine Translation Skills Among Iraqi Undergraduate Students

Post-editing machine translation often results in discrepancies between industry expectations and the actual quality of the final product. Despite the usually, low-quality outcomes associated with machine translation, it remains underutilized, even when there are opportunities for improvement through post-editing jobs. This study focuses on the impact of technology-assisted training on English proficiency among students. It investigates a standard error observed among Iraqi students during the post-editing of machine translations using Google Translate deciding whether to edit the machine translation output or to translate from scratch and understand the source text. This research involved 20 undergraduates from Al-Mustansiriyah University and employed a mixed methods approach, collecting data through an online questionnaire completed by students in the undergraduate translation program. The findings indicate a significant improvement in the frequency of post-editing skills among Iraqi students, serving as a distinguishing factor between the two groups. It also highlighted differences in translation abilities among students in Iraqi higher education, with notable improvements observed in the experimental group. In summary, this study supports previous research from the last five years, indicating a more favourable attitude toward machine translation and suggesting that post-editing is considered an essential sub-competency in translation education.

Systematic Review of English/Arabic Machine Translation Postediting: Implications for AI Application in Translation Research and Pedagogy

The twenty-first century has witnessed an extensive evolution in translation practice thanks to the accelerated progress in machine translation tools and software. With the increased scalability and availability of machine translation software empowered by artificial intelligence, translation students and practitioners have continued to show an unwavering reliance on automatic translation systems. Academically, there is little recognition of the need to develop machine translation skillsets amongst translation learners in English/Arabic translation programs. This study provides a systematic review of machine translation postediting with reference to English/Arabic machine translation. Using the Preferred Reporting Items for Systematic Review and Meta-Analysis, the paper reviewed 60 studies conducted since the beginning of the twenty-first century and classified them by different metrics to identify relevant trends and research gaps. The results showed that research on the topic has been primarily prescriptive, concentrating on evaluating and developing machine translation software while neglecting aspects related to translators’ skillsets and competencies. The paper highlights the significance of postediting as an important digital literacy to be developed among Arabic translation students and the need to bridge the existing research and pedagogic gap in MT education.

Translating Ulyssei@s: Work in progress between general and specialised translation in light of NMT

It is common knowledge that the publication of translation projects carried out by students is a very motivating factor for training. Since the second semester of 2015-2016, undergraduate translation students of the Faculty of Arts and Humanities of the University of Porto have participated in the development of the French and English versions of the digital encyclopaedia Ulyssei@s, created by the Institute for Comparative Literature Margarida Losa (ILCML) in the early 2010s. Besides motivating students, this translation project published on the Internet allows us to develop translation skills, foster teamwork, consolidate the three-phase methodology proposed by Gouadec (2002, 2007), and experiment with several essential tasks in language services: collaborative glossary-making, translation, revision of human translation and post-editing neural machine translation (NMT). Some of the challenges posed by the translation of Ulyssei@s are common to all types of translation, while others are specific to the translation of entries of this sui generis encyclopaedia, such as a variety of textual genres (biography, literary criticism, prose, poetry, theatre, essay) in hybrid texts and the need to find the original or the official translation of the works cited. This paper focuses on the data collected in the spring semester of 2022 and is based on the results of a survey submitted to students at the end of the project. It aims to share the experience of developing an inevitably unfinished project and adapting training and learning to the latest advances in Human Language Technology, since NMT has become a valuable tool for human translators, and Ulyssei@s is a digital encyclopaedia regularly enriched by ILCML collaborators with new entries.

The quality and productivity of post-edited machine translation in comparison with human translation : A case of Korean-English legal translation

The quality and productivity of post-edited machine translation in comparison with human translation : A case of Korean-English legal translation

Qualität des Posteditierens von maschinell erstellten Übersetzungen technischer Texte aus dem Slowakischen ins Deutsche durch erstsprachig slowakische Posteditoren

Summary While the rule of thumb that translators translate into their first language (L1) is often desirable, the demand for translations from lesser used into widely used languages often forces translators into this practice because it outstrips supply. The aim of this study is to show readers how the quality of the reference human translation (REF) produced by translators with Slovak as their L1 can influence the assessment of the quality of machine translation output (MT output), using the example of the light post-edited machine translation output (MTPE output) of technical documentation (TD) from Slovak into German by post-editors (PE) with Slovak as their L1. We therefore use the percentage difference between the TER and HTER scores to compare those segments of the studied TD that the machine translated differently than the human, whereby the PE nonetheless adopted such MT output or only partially changed it. We have divided the results into 6 categories of segments based on the differences between TER and HTER, where the difference between TER and HTER was below 50 %. We found that inappropriate REF and different, but not always wrong, solutions adopted by the PE compared to the REF translator can also result in poor evaluation of the MTPE output.

Improving machine translation and post-editing for Chinese tourism texts using transformer-based models

As the digital age and globalization continue to evolve, the demand for accurate machine translation of tourism texts has increased substantially. This paper investigates how to improve the quality of machine translation (MT) and machine translation post-editing (MTPE) of Chinese tourism texts for non-native speakers. A review of the machine translation literature reveals a significant progression in translation methods from rule-based to corpus-based, statistical, and finally to the current neural machine translation (NMT) models. Despite its advanced capabilities, NMT requires large amounts of parallel data for training, which often presents challenges. This study proposes the use of Transformer-based models for MT and MTPE to improve translation quality. A dataset was curated from online sources, mainly Chinese tourism websites. The methodology involved pre-processing the data, performing machine translation using the Transformer model, and post-editing the results. The experiment demonstrated an increase in the BLEU score, suggesting an improvement in translation quality. However, challenges such as the handling of synonyms and geographical nouns were encountered, indicating the need for further research and model optimization.

Student Performance in Machine Translation Training: Assessment and Exploration

The present research aims to conduct a process-oriented analysis to measure whether a group of graduate students enrolled in a translation course made steady progress in their performance of identifying machine translation (MT) errors and post-editing MT drafts of company web texts and news texts. A mixed methods approach consisting of quantitative and qualitative analyses was used. The findings show that there was a steady decline in the average number of MT errors that students could not spot or correctly identify in their three assignments. However, there was no significant improvement in student MTPE performance, with only a slight decrease in errors in the final MTPE assignment, which still remained worse than the first one. Finally, student responses in their reflection essays indicated that their reception of MT and MTPE had shifted from negative denial to positive acceptance. Overall, the findings of the present study reveal the need to extend the period of MTPE training for students. Incorporating MT training into the translation course has proven to be worthwhile for students, as it helps to dispel students’ previous misconception about MT and MTPE.

Re-thinking Machine Translation Post-Editing Guidelines

Machine Translation Post-Editing (MTPE) is a challenging task. It frequently creates tension between what the industry expects in terms of quality and what translators are willing to deliver as an end product. Conventional approaches to MTPE take as a point of departure the distinction between light and full MPTE, but the division gets blurred when implemented in an actual MTPE project where translators find difficulties in differentiating between essential and preferential changes. At the time MTPE guidelines were designed, the role of the human translator in the MT process was perceived as ancillary, a view inherited from the first days of MT research aiming at the so-called Fully Automatic High Quality Machine Translation (FAHQMT). My proposal challenges the traditional division of MTPE levels and presents a new way of looking at MTPE guidelines. In view of the latest developments in neural machine translation and the higher quality level of its output, it is my contention that the traditional division of MTPE levels is no longer valid. In this contribution I advance a proposal for redefining MTPE guidelines in the framework of an ecosystem specifically designed for this purpose.

Impact of Fine-Tuned Machine Translation on Medical Documentation Translation and Client Satisfaction

Aim: The aim of this study was to assess the impact of Fine-Tuned Machine Translation (FTMT) models, followed by Machine Translation Post-Editing (MTPE), on the translation quality and client satisfaction in medical documentation. Materials and Methods: The research analyzed 733,632 words across 317 projects completed in 2023 by Lingrowth, a medical and life science translation service provider. These projects involved translations from 16 source languages to 34 target languages. Document types included Instructions for Use (IFU), Investigator Brochures (IB), Summary of Product Characteristics (SmPC), Clinical Study Reports (CSR), and Informed Consent Forms (ICF). The projects were categorized into two groups for comparison: translations performed by human translators using Translation Memory (TM) and those processed by FTMT followed by MTPE. The primary metrics for evaluation were the number of post-project finalization requests and instances of severe negative feedback from clients. Results: It was observed that human-translated projects had a higher frequency of post-project finalization requests, whereas MTPE projects after FTMT exhibited a marginally higher rate of severe negative feedback. However, statistical analysis indicated that these differences were not significant, suggesting that the introduction of FTMT models with subsequent MTPE does not adversely affect the overall quality of medical document translations. Conclusion: The study concludes that FTMT models, when supplemented by MTPE, are com-parable in effectiveness to traditional human translations in the context of medical documentation. This highlights the potential of integrating FTMT in translation workflows without compromising translation quality or client satisfaction.

The Task of Post-Editing Machine Translation for the Low-Resource Language

In recent years, machine translation has made significant advancements; however, its effectiveness can vary widely depending on the language pair. Languages with limited resources, such as Kazakh, Uzbek, Kalmyk, Tatar, and others, often encounter challenges in achieving high-quality machine translations. Kazakh is an agglutinative language with complex morphology, making it a low-resource language. This article addresses the task of post-editing machine translation for the Kazakh language. The research begins by discussing the history and evolution of machine translation and how it has developed to meet the unique needs of languages with limited resources. The research resulted in the development of a machine translation post-editing system. The system utilizes modern machine learning methods, starting with neural machine translation using the BRNN model in the initial post-editing stage. Subsequently, the transformer model is applied to further edit the text. Complex structural and grammatical forms are processed, and abbreviations are replaced. Practical experiments were conducted on various texts: news publications, legislative documents, IT sphere, etc. This article serves as a valuable resource for researchers and practitioners in the field of machine translation, shedding light on effective post-editing strategies to enhance translation quality, particularly in scenarios involving languages with limited resources such as Kazakh and Uzbek. The obtained results were tested and evaluated using specialized metrics—BLEU, TER, and WER.

In search of a fair MTPE pricing model: LSPs’ reflections and the implications for translators

ABSTRACT Machine translation post-editing (MTPE) has become an important part of the commercial translation business. As such, MTPE pricing models are under-studied yet vital elements in determining the professional value of translators as workers. This focus-group study investigates the discourse of eight LSP representatives relating to their MTPE pricing practice and, using the notion of psychological contract from organisational psychology, analyses what kind of external messages, including untold meanings and cues, translators may receive from the MTPE pricing practice. We then discuss how such messages may shape translators’ individual beliefs relating to contractual terms in MTPE projects, as well as their implications for MTPE operations in general. Three significant points are identified regarding; (1) How LSPs use different pricing models for different stakeholders to manage translators’ motivation for MTPE work; (2) How the notion of ‘speed’ incorporated in the MTPE pricing methods has a potential to determine translators’ work motivation and recruitment; and (3) How the power structure of technology ownership may influence future MTPE pricing practice. The outcomes of this exploratory study will contribute to the development of much-needed industry-wide conversation about this controversial topic.

POST-EDITING DECISIONS IN MACHINE TRANSLATION OF LEGAL CONTRACTS

Legal translation requires special language use. It has therefore been many a time credited to be one of the most difficult genres to translate. Post-editing of machine translation is becoming more and widely employed in the translation industry. Even in languages that cannot boast huge amounts of resources, translating texts of certain genres is rarely conducted without resorting to machine translation, as the resulting quality seems to be approaching parity with human translation. This study attempts to analyze and compare post-editing decisions in machine translated legal contracts from English to Lithuanian and from English to Ukrainian. Legal contracts are part of the legal discourse with language characterized as being archaic, complex, culturally bound and typically stocked with formulaic phrases, specialized terms, and Latinisms. Being a technical type of discourse, legal translation may benefit from machine translation. The conclusions of the study are based on the findings of assessment of machine translation post-editing decisions made by professional translators. The findings demonstrate and support the results of previous research with other languages that the main trends are as follows: changes in noun and verb phrases in both languages, co- reference and reordering changes. Further study of the stated problem will allow expanding (targeting and detailing) the typology of post-editing actions, considering the specifics of Baltic (Lithuanian) and Slavic (Ukrainian) languages.

AI аnd Machine Translation Post-editing: Advancements and Challenges (Insights for Students of International Studies)

Now, with the appearance of the latest wave of a sophisticated generative artificial intelligence (AI), humanity is about to embark on an entirely new functioning order. The challenge today is that due to AI the world will definitely undergo drastic metamorphosis tomorrow, and again the day after. To adapt to this reinvented economy, people will need to reinvent their skills, careers – and, indeed, their lives. Therefore, educating people for reinvention in this fluid context will require the reinvention of higher education itself. According to Joseph E. Aoun, taking into consideration the targets to be met in the nearest future, the next generation of HEI students are supposed to be educated to invent, to create, and to discover – to meet society’s targets that any most sophisticated artificial intelligence agent cannot, consequently, a curriculum should include technological literacy, or understanding how machines work and how to work with them. Furthermore, the emergence of cognitive translation studies has stipulated an interdisciplinary approach to delve into the cognitive and behavioural aspects of a broad array of cross-language activities including all kinds of translation and interpreting. In a world that relentlessly pursues efficiency and productivity, the figure of a post-editor, a professional translator who has the skills to add that necessary human touch to a text which has previously been subjected to software algorithms, has become more prominent.

Performance and perception: machine translation post-editing in Chinese-English news translation by novice translators

Machine translation has become a popular option for news circulation, due to its speed, cost-effectiveness and improving quality. However, it still remains uncertain whether machine translation is effective in helping novice translators in news translation. To investigate the effectiveness of machine translation, this study conducted a Chinese-English news translation test to compare the performance and perception of translation learners in machine translation post-editing and manual translation. The findings suggest that it is challenging for machine translation to understand cultural and semantic nuances in the source language, and produce coherent structural translation in the target language. No significant quality difference was observed in post-editing and manual translation, though post-editing quality was found to be slightly better. Machine translation can help to reduce translation learners’ processing time and mental workload. Compared to manual translation, machine translation post-editing is considered as a preferred approach by translation learners in news translation. It is hoped that this study could cast light on the integration of machine translation into translator training programs.

Scaling Students’ Self-Efficacy on Machine Translation Post-Editing: Psychometric Properties of the Scale and Their Associations

Machine Translation Post-Editing (MTPE) has emerged as a productivity-enhancing practice in the language service industry, where human editors correct the output of machine translation systems. To ensure that students of translation possess the necessary skills for MTPE, it is essential to understand their self-efficacy in this domain. This research paper aims to assess students' self-efficacy in translation learning, specifically in the context of MTPE, and explore the factor structure, psychometric properties, and internal associations of their self-efficacy. The study utilized a modified survey adapted from the Scale for Assessing Translators' Self-Efficacy and collected responses from 65 undergraduate students in a Chinese university. The survey data underwent reliability and validity analyses, including exploratory factor analysis, to assess the measurement tool's consistency, stability, and construct validity. The results indicated a high reliability of the scale (Cronbach's Alpha = 0.914) and revealed three primary dimensions of self-efficacy: Decision-making of MTPE, Communicative Competence of MTPE, and Strategic Competence of MTPE, and the strong inter-correlations suggests that they collectively measure the construct of translators' self-efficacy of MTPE, providing insights into the skills and abilities required for effective MTPE. The findings contribute to the development of psychometric tools for further research in translation and promote pedagogical reform to align with evolving market trends emphasizing human-machine collaborative translation.

Maschinelle, posteditierte und menschliche Übersetzung publizistischer und populärwissenschaftlicher Texte aus dem Slowakischen ins Deutsche

Abstract Machine translation from small into big languages, in our case a transfer from Slovak into German, has not yet undergone thorough research. Our goal is an analysis and comparison of machine translation (MT) and post-edited machine translation (PEMT) errors with human translation (HT) as well as detecting the efficiency rate of post-editing within that mentioned comparison. Hereby we apply a method of automatic and manual metrics by percentages and evaluation numbers. Results indicate that post-editing is more efficient which however does not mean that human translation has a lower quality. A higher degree of freedom of HT allows for more creative solutions contrasting with post-edited MT in the results.

Post-Editing Machine Translation (PEMT) as the preference method for university students in Indonesia

Language translating process by Indonesian students is particularly suffered from many errors. Its defined strategy should be carefully watched due to completing its translating perfection. Thus, this study identifies the problematic translation ways commonly referred and used by university students in Indonesia. The mixed method using embedded design between qualitative and quantitative data with a case study approach was carried out to find the preferences of the translation method used by the students. The respondents are 50 English Department students from different levels of studying years. The result shows that Post-Editing Machine Translation (PEMT) was preferred among other translating methods. The reasons are related to the ease and quality of the translation work. 76.90% of the students responded and chose to have PEMT as the preference. However, there were only 53.8% of them implementing PEMT as the translating process, and 38.5% were familiar with Machine Translation (MT) without post-editing instead. It indicates an inconsistency in the use of the translation methods as each of them experienced and chose different type of methods. However, PEMT was placed as the most referred translating method used by the students. This suggests that the presence of technology is well utilized while still paying attention to the quality of the translation through the post-editing process.

Comparative Analysis of Post-editing Techniques for Chinese-to-English Translation Tasks: A Quasi-Experimental Study

This study aimed to examine the differences in translation time and quality between human translation (HT), machine translation post-editing (MTPE), and ChatGPT post-editing (ChatGPT PE). A quasi-experimental design was employed, involving 30 junior participants majoring in English from a Chinese university. A translation task consisting of 532 words from Chinese to English was assigned to the three groups and the translation speed and quality were assessed using the Multidimensional Quality Metrics (MQM) and Dynamic Quality Framework (DQF). The results demonstrated that the post-editing technique could produce translation quality comparable to human translation while achieving faster translation speed. It was observed that ChatGPT PE outputs exhibited the highest number of terminology errors but had the fewest accuracy errors, while mean differences in two of accuracy dimensions were observed among the three groups. The utilization of post-editing in language teaching would allow for the integration of machine translation strengths and human expertise, resulting in high-quality translations on par with human translations. To optimize post-editing outcomes, it is crucial to develop and teach error-detection techniques and avoid excessive reliance on AI technology.

Effect of text type on translation effort in human translation and neural machine translation post-editing processes: evidence from eye-tracking and keyboard-logging

ABSTRACT Text type influences the translator’s decision in taking an overall translation approach and consequently selecting translation techniques. While the majority of studies to date have investigated this phenomenon as an end-product, the effect of text types on the translation process has received little attention in the area of cognitive translation processes, especially from an empirical perspective. To fill this gap, the current study employed eye-tracking and keyboard-logging approaches to investigate the effect of text type (informative text, expressive text, and operative text) on the cognitive efforts demanded in the Human Translation (HT) and Neural Machine Translation Post-Editing (PE) processes. The data was collected from 24 participants (12 professional translators and 12 undergraduate students), who were required to manually translate and post-edit three text types from Chinese into English. From the findings, it appears that text types affect the translator’s time investment, mental process, and technical input in both HT and PE processes. The findings also suggest that PE, as one mode of Human–Computer Interaction, emancipates translators from the effortful task of traditional from-scratch translation, and this effort change is also affected by the text type. The study has practical implications for both translators and translator training.

Post-editing of Machine Translation of An English-into- Arabic Text by Translation Students

Post-editing of Machine Translation of An English-into- Arabic Text by Translation Students