Team dynamics and efficiency in post-editing: Machine translation systems versus artificial intelligence chatbots under tight deadline conditions

Team dynamics and efficiency in post-editing: Machine translation systems versus artificial intelligence chatbots under tight deadline conditions | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Team dynamics and efficiency in post-editing: Machine translation systems versus artificial intelligence chatbots under tight deadline conditions Honghui Hu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6892673/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This research investigates the post-editing (PE) of machine- and artificial intelligence (AI)-generated technical documents in the context of tight deadline jobs (TDJs) to assess team dynamics and efficiency. In a mock translation agency setting, 12 trainee translators were divided into two groups: Team 1 focused on PE Baidu’s machine translation (MT), while Team 2 worked on the outputs from Kimi AI, following a convergent parallel mixed-methods design. Quantitative metrics monitored time, errors, and consulted resources, while qualitative insights explored stress levels and teamwork. Results indicate that Team 1 invested more time (60 vs. 48 minutes) and corrected more errors (19 vs. 12) due to the prevalence of mistakes in Baidu’s output. They also consulted more resources but achieved lower accuracy (92% vs. 95%) than Team 2, which improved Kimi AI’s fluent yet contextually flawed translations. Both teams tended to prioritize speed over quality, with ongoing challenges in terminology. The study's implications for the professional approach to translator training (PATT) involve simulating AI/MT PE to improve error detection, making contextual adjustments, incorporating stress management strategies, and promoting terminology proficiency and flexible team roles. This research helps address gaps in PATT related to TDJ training, machine-translated post-editing, AI functionality integration, and specialized text training, thereby enhancing translator education to align with current industry needs. Humanities/Language and linguistics Social science/Education Social science/Language and linguistics Social science/Social policy Social science/Sociology Machine translation post-editing AI translation Tight deadline job Professional approach to translator training Team dynamics Translator training Figures Figure 1 1. Introduction The translation industry is undergoing a significant transformation influenced by artificial intelligence (AI) and automation, necessitating training that meets market demands and professional standards. Rapid developments in machine translation (MT) and AI chatbots have transformed workflows, making the post-editing (PE) of AI- and MT-generated texts a core aspect of professional practice (Bowker & Buitrago Ciro, 2019). Tight deadline jobs (TDJs) are becoming more prevalent, necessitating efficiency, collaboration, and the ability to adapt under pressure. However, conventional translator training often falls short, emphasizing linguistic accuracy over the practical, technology-based skills essential in global markets (Pym, 2009). This gap highlights the urgent need for curricula that replicate real-world conditions, incorporating AI tools, machine-translated post-editing (MTPE), and collaborative processes to equip graduates with the skills demanded by industry. A professional approach to translator training (PATT) meets this need by simulating professional workflows through project-based learning (PBL) and workplace scenarios (Olvera Lobo et al., 2007). It prioritizes learner-centered education and promotes translation competence through roles such as translator, terminologist, and project manager, utilizing tools such as Translation Memory systems and collaborative platforms (e.g., Aula.int, Basic Support for Cooperative Work [BSCW]). However, PATT encounters challenges, including a limited emphasis on TDJs, inadequate integration of MTPE, and a lack of exploration into AI chatbots or strategies specific to certain text types (Konttinen et al., 2019; Morón & Calvo, 2018). These limitations hinder its ability to adequately prepare students for high-pressure, technology-driven settings, underscoring the need for research that bridges academic training with industry requirements. This study examines the impact of PE machine- and AI-translated technical texts in TDJ conditions, with a focus on process efficiency and team dynamics. It addresses two research questions: (1) What impact do machine- and AI-translated texts have on the PE process in a time-sensitive professional environment? (2) How do TDJ conditions affect collaborative team interactions? Conducted in a simulated translation training setting with 12 trainee translators divided into two teams, one PE a Baidu MT output and the other a Kimi AI output, the study simulates a real-world scenario involving a technical news article. The design meets industry requirements for swift and precise translations. Key terms in this study consist of the following: team cohesion, which refers to how unified and supported trainees felt, reflecting their collaboration and team spirit; challenges encountered, representing obstacles such as technical difficulties or time limitations; communication, which encompasses the exchange of ideas, including both initial planning and ongoing discussions; TDJ perceptions, indicating trainees’ experiences of stress and pressure within tight timeframes; and team organization, highlighting the arrangement of roles and tasks to effectively manage the project. This study enhances the PATT framework by tackling essential gaps in TDJ preparation, MTPE, AI integration, and training tailored to specific text types. Exploring the connections between technology and teamwork in high-pressure situations reveals how translator training can be enhanced to effectively leverage contemporary workflows. The results influence curriculum development by promoting stress management techniques, collaborative tools, and terminology strategies, all aimed at boosting employability. Furthermore, this research advances translation pedagogy by providing empirical support for AI and MT PE practices, aligning with market demands and equipping graduates for the evolving, AI-centered translation industry (Pym, 2009; Olvera Lobo et al., 2007). 2. Literature Review The literature review section focuses on professional translation training. 2.1. Professional Approach to Translator Training PATT is a cutting-edge approach to translation instruction that emulates real-world professional environments by assigning students specific roles in a PBL setting (Olvera Lobo et al., 2007). This method incorporates information and communication technology to simulate a translation agency. In a PATT instructional environment, trainees assume various roles within translation teams, engaging in tasks that typically involve documentation, terminology management, translation, revision, and formatting. These activities develop essential skills in teleworking, teamwork, and interdisciplinary collaboration (Alvera Lobo et al., 2007). The process reflects a professional translation workflow, encompassing pre-translation (research and terminology), information storage, linguistic conversion (representation), and delivery of the final text (communication) (Olvera Lobo et al., 2005). PATT prioritizes learner-centered education, resonating with the Bologna Declaration’s emphasis on career-oriented teaching and promoting student autonomy (Olvera Lobo et al., 2007). This approach enhances translation competence by integrating communicative, extra-linguistic, instrumental-professional, and strategic sub-competences, thereby embedding practical, market-relevant skills within the curriculum. A significant contribution to PATT was made at the University of Granada, Spain (Olvera Lobo et al., 2007) as part of its four-year Translation and Interpreting degree program. This was implemented using the Aula.int platform, enhanced by the BSCW system. This setup enabled asynchronous collaboration, allowing students to organize tasks, share documents, and communicate online, effectively preparing them for the teleworking aspects of the translation profession. The approach provided numerous advantages by simulating a translation agency through the Aula.int platform. It immerses students in a realistic professional setting, assigning roles such as translator, terminologist, and project manager, thereby fostering practical skills in documentation, terminology, and formatting (Olvera Lobo et al., 2007). PATT enhances both teleworking and teamwork via the BSCW platform, which facilitates asynchronous collaboration and prepares students for the global market. It combines interdisciplinary skills from various course modules, encouraging a comprehensive understanding of translation competence, which includes communicative and instrumental-professional sub-competences (PACTE, 2000). By promoting student autonomy and self-directed learning, PATT increases engagement and motivation while addressing technical challenges, effectively equipping students for the demands of the translation profession. 2.2. Related Studies on PATT Studies employing a professional translation approach have been theoretical, practical, or a combination of both. Pym (2009) is one of the most theoretically focused scholars advocating for training that meets job demands, particularly in the digital era. In his 2009 work, he critiqued traditional translator training for its instructional, teacher-centered nature, where students translate primarily for assessment against literal models by often non-professional teachers, which fails to develop professional competencies. He contended that this transmissionist model overlooked market requirements, leading to graduates who are ill-prepared for industry expectations (Bowker, 2004; Gouadec, 2007). Pym urged that training should reflect real-world practice through authentic tasks, portfolios, and social constructivist approaches, enabling students to collaboratively build skills such as problem-solving and client engagement (Kiraly, 2000). He also emphasized the importance of incorporating translation memories and specialized software (e.g., localization tools) to align with industry standards and specific sectors, highlighting professional strategies that correspond to actual competencies (Künzli, 2004; Kussmaul, 1995). Beyond theoretical frameworks, numerous empirical studies have explored the alignment of training with practice through various methods. In the PBL framework, researchers have emphasized the importance of student-centered, constructivist strategies in translator education, focusing on authentic, collaborative, and technology-oriented tasks. Mitchell-Schuitevoerder (2013) examined the role of reflective group work and blogs in fostering critical thinking within a translation technology module. Morón and Calvo (2018) incorporated transcreation through simulated marketing projects to address intercultural issues, while Li et al. (2015) employed PBL in business translation, implementing real-world projects to enhance practical skills. García González and Veiga Díaz (2015) adopted Guided Inquiry and peer reviews in specialized translation classes to align with higher education objectives. Tekwa et al. (2024) integrated instant messaging into collaborative tasks within a project-based translation and localization training context, while Apandi and Afiah (2019) centered on cultural translation, promoting independent learning. Research indicates that PBL promotes professional skills, employability, and critical thinking through authentic tasks, such as transcreation and localization (Morón & Calvo, 2018; Tekwa et al., 2024). It encourages collaboration and engagement, evidenced by positive student feedback (Li et al., 2015; Mitchell-Schuitevoerder, 2013). However, some drawbacks include limited preparation time, difficulties in group adaptation, and logistical challenges, such as accessing observation sites (Apandi & Afiah, 2019). The translation of cultural content and the integration of new tools, such as transcreation or instant messaging, present complexities and often lack established frameworks or quantitative validation (Morón & Calvo, 2018; Li et al., 2015). These challenges underscore the need for enhanced resources and research to improve the effectiveness of PBL in translator training. Alongside PBL, PATT researchers have implemented a workplace simulation approach, focusing primarily on workflows at translation companies, authentic project-based tasks, and translator workstations (Konttinen et al., 2019; Mo & Man, 2017; Prieto-Velasco & Fuentes-Luque, 2018). This approach incorporates real-world conditions, MT-assisted translation memory (TM) tasks, and role-based scenarios, such as those involving translation specialists or project managers (Bundgaard, 2017; Kozyar et al., 2022). The tools employed include TM systems, MT, Web 2.0 collaborative platforms, terminology databases, and project management software (Buysschaert et al., 2018). Some studies emphasize the holistic integration of various technologies, while others focus on standard computer-assisted translation (CAT) tools and online resources, which enable students to engage with industry-standard workflows and authentic translation assignments in simulated environments (González-Davies & Raído, 2018). The simulations have aimed at developing professional translation competence, foreign language proficiency, and technology-based skills, such as using CAT tools and Web 2.0 platforms (Mo & Man, 2017; Prieto-Velasco & Fuentes-Luque, 2018). Organizational, interpersonal, and problem-solving skills have also been fostered, alongside confidence and autonomy (Konttinen et al., 2019; Kozyar et al., 2022). Benefits include enhanced employability through industry-aligned training, improved academic performance, and positive attitudes toward collaborative tools (Buysschaert et al., 2018). Simulations bridge the gap between academic training and professional practice, preparing students for real-world pressures and enabling them to adapt to technological imperfections, ultimately enhancing their ability to think and act like professional translators (Bundgaard, 2017; González-Davies & Raído, 2018). 2.3. Existing Lacunae and Research Questions Research on PBL and simulated workplace environments in the PATT framework uncovers several significant gaps that hinder their relevance to current translation practices. First, both PBL and simulation research largely neglect TDJs, which are prevalent in professional environments and demand swift decision-making and efficiency under pressure. Although PBL emphasizes collaboration (e.g., Mitchell-Schuitevoerder, 2013; Tekwa et al., 2024), it overlooks how collaborative dynamics operate under the time constraints of TDJs, creating a gap in preparing students for high-pressure environments. Furthermore, neither PBL nor simulations under PATT sufficiently investigate MTPE, despite its prominence in contemporary translation workflows (Bundgaard, 2017 mentions MT-assisted TM but does not specifically address PE). The role of AI, especially AI chatbots, is also missing, despite their increasing involvement in translation tasks for terminology assistance and drafting. Finally, the importance of text types, which are essential for customizing translation strategies, is not a focal point in PATT research, despite workplace simulations encompassing diverse tasks (e.g., transcreation, localization) (Konttinen et al., 2019; Morón & Calvo, 2018;). These gaps highlight the need for PATT to incorporate TDJs, MTPE, AI tools, and text-type-specific strategies to better reflect the evolving demands of the translation industry and enhance graduate employability. This study addresses these gaps by examining the PE of a technical text under strict deadline conditions. It replicates a work environment with two trainee groups: one that post-edits a machine-translated text and another that post-edits an AI-translated text. The research addresses the following research questions: How do machine- and AI-translated texts affect the PE process under tight deadline conditions? How do TDJ conditions impact team performance dynamics? This study significantly enhances the PATT framework by tackling major gaps in current PBL and simulated workplace studies, especially concerning TDJs, MTPE, AI integration, and team dynamics. 3. Methodology The approach utilized involved simulating actual practices in the translation industry. The client, a news media company, supplied an 836-character text that needed to be posted online within two hours, requesting an urgent translation. A group of 12 trainee translators was created through simple random sampling and divided into two teams of six. Each team translated the source text (ST) using different tools. Team 1 employed MT systems for their translation, which they subsequently post-edited. In contrast, Team 6 utilized AI chatbots for the same task, followed by PE. This methodology aimed to replicate real-life scenarios faced by translators who sometimes need to deliver translations on tight deadlines, explore the effects of MTPE and AI chatbot PE (AIPE) processes, evaluate teamwork dynamics under these conditions, and collect students' perceptions of various translation tools. 3.1. Participants Participants included 12 trainee translators in their second semester of the MTI program at a Chinese university. This sample aligns with current research in the field (McDermid, 2025; Tsai, 2025), as graduate class sizes are typically small. All participants were enrolled full-time in the MTPE course, where they developed their translation and PE skills, collaborative teamwork, and understanding of the role of AI in facilitating translation. The class consisted of eight female and four male students who were admitted to the program after a careful selection process and an entrance examination that evaluated their translation and writing skills in both Chinese and English. In the MTPE course, they translated texts from Chinese to English in the CAT laboratory, a multimedia classroom specifically designed for translation instruction. Baseline scores were measured based on their performance in prior MTPE and AI-based MTPE activities that preceded this intervention. All ethical requirements were satisfied, and informed consent was obtained from the trainees. 3.2. Design The experiment was conducted in Week 14 of the 18-week semester and spanned 80 minutes. From Weeks 3 to 10, students focused on PE texts with different MT systems, including DeepL, Google Translate, Baidu Translate, and Yandex. In Weeks 11 and 12, trainees learned about AI prompting and how to utilize AI chatbots for PE. Specifically, they employed an AI chatbot to translate the text and then, using various prompts, conducted terminology research and enhanced the grammar of English sentence structures. In Week 13, both PE techniques (informed by MT and AI) were reviewed in class to ensure all students had a consistent baseline. They completed three assessments before the intervention—Week 4 and Week 10 for MTPE, and Week 12 for AIPE. The assessment in Week 10 was a group-based activity. No statistical differences were observed among the trainees. Moreover, teamwork was a key component of most in-class activities. Students routinely collaborated in teams, learning to allocate tasks, discuss ideas, and adopt various roles within translation scenarios. This pedagogical strategy aligns with PBL (Apandi & Afiah, 2019 ), a recognized method for simulating workplace environments that helps students gain insights into both the complex nature of the learning process and the translation process itself (Kiraly, 2005 ). The intervention took place in the CAT laboratory, where the two teams were positioned at opposite ends, separated by three workstations. This arrangement ensured that discussions held by one team did not interfere with those of the other. Trainees received the text and were informed that the client expected the translated version within 70 minutes. Subsequently, they were provided with instructions on handling the text, as detailed in Fig. 1 . Two trained assistants oversaw the experiment using a checklist to maintain consistency. Translated text Trainees translated a technical news piece about U.S. export regulations impacting the electronic design automation sector, aligning with the changing requirements of the translation field. This genre exemplifies the high-stakes, urgent nature of technical translation in the global tech industry, while also filling a gap in PATT research regarding TDJs and specific text-type strategies (Morón & Calvo, 2018). The intricate terminology and cross-disciplinary themes (technology, finance, geopolitics) made it suitable for assessing machine- and AI-translated text PE under pressure, and fostered the exploration of team dynamics. By focusing on a text that reflects real-world translation challenges, this study can enhance the employability of trainees within the PATT framework. Translation assessment The translation assessment methodology draws from the Multidimensional Quality Metrics (MQM), Post-Editing of Machine Translation (PEMT), and the LISA QA Model, emphasizing fluency, accuracy, and terminology as supported by Koponen ( 2010 ), Lommel & Burchardt (2014), and Park & Padó (2024) for MQM; O’Brien (2021) and Carl et al. ( 2015 ) for PEMT; and Hariyanto (2016) and Lommel (2018) for the LISA QA Model. The evaluation criteria encompassed fluency (spelling, collocation/cohesion, word form, punctuation, and stylistic issues such as subject-verb agreement, tense, missing articles, subject-verb complement, and adverb placement) and accuracy (omission, addition, incorrect term, untranslated word/expression, over-translation, under-translation, British vs. American English, and mistranslation). All errors were weighted equally to ensure uniformity and consistency throughout the evaluation process. Three assessors, collectively having over 50 years of experience in translator training, performed the assessment. They initially assessed the translation output independently before engaging in a collaborative discussion to align their scores, thereby ensuring a consistent and standardized evaluation. 3.3. Data Collection and Instruments The translation assessment’s data collection process utilized various tools to thoroughly evaluate the PE task. Students used a digital timer to record the start and end times, measuring time efficiency. The translation output was examined for accuracy and fluency, with students noting the resources they referred to, including dictionaries, websites, glossaries, and term banks. Furthermore, students submitted typed personal reflection notes discussing team cohesion, challenges, communication, their perceptions of the translation and PE processes, and team organization. Quantitative data, such as time efficiency metrics, were analyzed with the Excel Data Analysis tool, while qualitative data from reflection notes were assessed using NVivo to gauge their relevance in simulating a real-world workplace environment. 3.4. Data Analysis 3.4.1. Quantitative Data Analysis The differences in scores between the two teams were determined through the evaluation of the PE text. Assessments were made regarding both accuracy and fluency. Three key factors were taken into account: the number of correctly edited errors, unedited errors, and incorrect edits. The total score of 100 was adjusted by subtracting the counts of unedited and incorrect edits to derive the performance score for each team. Fluency errors were assigned a weight of 1, while adequacy errors, deemed more significant for meaning, received a weight of 2. The simplification of error weights aimed to improve scoring consistency. At times, a single error within one sentence could be interpreted in various ways, and in such instances, “the analysis with the lowest total error count was selected” (Koponen, 2010 , p. 6). Qualitative data: A deductive thematic analysis was conducted to assess the qualitative data obtained from trainees’ reflections. Identified themes included team cohesion, challenges faced, communication, particularly the discussions before starting the task, perceptions of TDJ, and team organization. The analysis utilized NVivo. Two trained coders meticulously reviewed the data and initially coded it using a predetermined set of codes (Appendix 1). After initial coding and discussion, the coders’ reliability was determined to be significantly high (Cohen’s kappa [k] = 0.87). Within the NVivo tool, the codes were examined in the Nodes section, and patterns, similarities, and connections were investigated by organizing them into parent–child relationships. This approach facilitated the extraction of insights on how the two teams operated, as well as their perceptions of the TDJ and the work environment simulation. 4. Findings 4.1. Quantitative Findings 4.1.1. Readability Scores Kimi's translation had a Flesch Reading Ease score of 42, making it slightly easier to read than Baidu's score of 39.3. However, both are categorized as “difficult” and are suitable for educated readers. In terms of complexity, Baidu’s translation recorded a Flesch–Kincaid grade level of 13.3 and a Gunning Fog index of 15.5, in contrast to Kimi’s scores of 12.7 and 15.2, respectively. Kimi’s version was lengthier, comprising 574 words and 27 sentences, whereas Baidu’s contained 444 words and 20 sentences, despite Baidu’s sentences being slightly longer on average (22.20 vs. 21.26 words). Both translations featured 12 paragraphs, and their percentage of complex words was similar (Kimi: 19.16%, Baidu: 19.4%). However, Kimi used a greater number of complex words in total, attributable to its higher word count, as illustrated in Table 1. Table 1. Differences between the Baidu and Kimi translations 4.1.2. Error Types Our error type analysis indicated that Kimi AI outperformed Baidu Translate in terms of both fluency and accuracy. In terms of fluency, Kimi AI demonstrated better grammatical consistency, including correct tense and subject–verb agreement, as well as more natural phrasing, avoiding awkward calques such as Baidu’s “platformization.” However, both tools had issues with adverb placement and missing articles. Regarding accuracy, Kimi AI made fewer critical mistakes, notably avoiding additions (e.g., the term “potential” in “US supply cut”) and over-translations; however, it shared some weaknesses with Baidu concerning terminology (e.g., “Gai Lun” vs. “Grain-Physics Electronics”) and untranslated institutional names (e.g., BIS). Baidu’s literal translations (e.g., “market has voted”) and omissions (e.g., leaving out the 2025 revenue clause) significantly affected clarity. In conclusion, while Kimi AI produced more idiomatic English, both translations still needed human review for technical terms and nuanced expressions to meet professional-grade quality. 4.1.3. Time and Resources Consulted Regarding the time spent on the task, Team 2 took 60 minutes to post-edit the text, while Team 1 took 48 minutes, which is 12 minutes less. This means Team 1 had more time to proofread and revise the post-edited text before submission. Meanwhile, Team 1 consulted significantly more resources than Team 2 during the PE task. As outlined in Table 2, Team 1 referred to eight resources, including online dictionaries, glossaries, term banks, corpora, social media, and websites. In contrast, Team 2 relied primarily on websites while significantly depending on the AI chatbot for terminological, grammatical, and stylistic references. Table 2. Resources consulted by the two teams 4.1.4. Performance Both teams encountered various fluency and accuracy errors that needed correction. Team 1, which worked on the Baidu MT text, discovered and addressed 19 errors, whereas Team 2, which focused on the AI-translated text, found and corrected 12 errors. The examples presented in Table 3 (refer to Appendix 2 for the complete list of errors) show that although some errors were the same, many others stemmed from the translation tool. Table 3. Examples of errors found and fixed by teams Since neither team post-edited the same text, we calculated the number of unidentified errors and subtracted that from a total of 100 points. Based on this calculation, Team 1 scored 92% and Team 2 scored 95%. The errors included fluency issues, such as incorrect use of passive voice, punctuation mistakes, and problems with sentence structure. 4.2. Qualitative Findings Data analysis highlighted both similarities and differences between the teams, facilitating a better understanding. Perceptual similarities spanned the four themes. While all groups shared common perceptions of TDJs, similarities in team cohesion, organization, and communication were noted before the text translation was conducted using the Baidu Translate MT system and the Kimi AI chatbot. 4.2.1. Perceptual Similarities Pervasive nature of the TDJ: Throughout both teams, the looming pressure of the strict deadline was a prominent factor, rated between 9 and 10 on a scale of 1–10. Both teams experienced significant stress and pressure due to the quick turnaround demands. This tight deadline infiltrated their workflow from the beginning, impacting their quality standards, decision-making processes, and overall well-being. The qualitative data suggested it was an underlying factor with a noticeable impact. Selections from the trainees’ reflection notes support our analysis. Excerpts from trainees: Team 1 Trainee 3 (T1T3) maintained that “the timer was like an enemy to us. Everyone minute was precious, and because the time kept ticking, we felt anxious. I was sure we would finish the job, but I doubted the quality of the output.” Additionally, T1T1 maintained: From the time we received the text, it was clear this would be a scramble. The tension was in the air, and we just wanted to start immediately. Because we wanted to save every second, I believe our first discussion was not well coordinated. Meanwhile, T1T4 expressed how the urgent deadline influenced their thinking, saying, I don’t think my mind was on doing a perfect job most of the time. I was rather focused on how we could finish the task within the time limit. I think this thought shifted my focus a bit and influenced my concentration during the PE process. T1T6 corroborated this perception, stating that the anxiety of not finishing in time and not producing the quality the client expected were consistent throughout the entire activity. “It felt like an examination that I had to finish in time and also obtain a high score.” Team 2 trainees had similar perceptions. T2T4 remarked, Translating the text was like a sprint. I was always looking at the time, thinking “Can we really polish this enough.” The pressure, for me, was to do a good translation; I don’t think I was thinking about a perfect translation. T2T1 added that they spent a significant amount of time during the initial discussion talking about the tight deadline; “luckily, our team leader asked us to try and focus on the task, which, was still a little hard to do.” Meanwhile, T2T5 explained that right up until the end, the worry of not producing a good-quality text persisted in their mind: The translation of Kimi AI was, to be honest, quite good. That gave us a good start, I think. But, even when we submitted the text, I was still worried we did not have enough time to produce a really good quality translation. Initial communication and team organization: Before translating and ultimately subjecting the texts to PE, both teams recognized the critical need for upfront planning, clear role allocation, and effective communication to manage the tight deadline. Their reflections revealed similar, proactive strategies in this crucial initial phase, which established a foundation for their work. T1T2 maintained: We had a quick huddle to divide the work. I think, as we have done in class several times, we needed a solid plan to avoid getting lost in the chaos of the deadline. So, I was responsible for the terminology work, since I can look up the dictionary faster and I type faster than most of my group members. Most Team 1 and 2 members shared the same ideas, focusing on how the work was divided based on individual team member qualities and skills. For example, T2T6 was responsible for terminology research because they could type quickly and mine information more efficiently: “I knew they would assign me to look up words and quickly update the target text. I am better at this than most of my team members.” Two types of organization were highlighted: One based on the individual skills of members and the other on sharing parts of the text among team members. As T1T5 maintained, “Our first ten minutes were all about who does what. We split the text into manageable chunks and assigned them. There was no time for indecision.” T2T3 confirmed this organization, noting, “Our first step was to assign sections and set internal mini-deadlines. We knew that without a tight organizational structure, the TDJ would overwhelm us. Communication at this stage was vital for alignment.” At this stage, communication was brief for both teams, essentially during the first ten minutes. The findings underscore the widespread stress caused by TDJ on two translation teams, affecting their workflow, decision-making, and overall well-being. Trainees expressed anxiety about juggling speed and quality, frequently prioritizing completion over perfection. Effective planning and skill-based role assignment helped alleviate the pressure. These results are consistent with Lazarus and Folkman’s (1984) stress research, which indicates that time pressure diminishes cognitive focus, as well as with Chesterman’s (2005) observations regarding translators’ focus on efficiency. Cognitive load theory (Sweller, 1988) elucidates the trainees’ compromises in quality, while Hackman’s (2002) research on team dynamics validates their successful role allocations. Initial communications reflect the findings of Marks et al. (2001) on the importance of concise coordination in high-pressure environments, highlighting common challenges faced in time-sensitive tasks. 4.2.2. Perceptual Differences Divergent experiences—post-translation dynamics: Once the initial machine-generated or AI-generated translations were available, the unique characteristics of these outputs resulted in significant differences in the challenges faced, team cohesion, and communication patterns that followed. Challenges encountered: The error types and the required PE effort varied significantly, leading to distinct challenges for each team. Team 1 trainees frequently highlighted issues stemming from the raw, often literal, and grammatically inconsistent MT output. For example, T1T1 maintained that “The machine-translated text was raw. I think Baidu did not get most words right, so we spent a lot of time fixing grammatical errors and awkward phrasing. To me, this felt like intensive editing rather than translation.” Four members of Team 1 shared this perception. T1T6 added that “Sometimes the MT completely missed the point or used the wrong terminology. That slowed us down because we had to assign more people to do terminology work rather than post-edit.” Meanwhile, T1T2, initially assigned with the terminology task, remarked, I think terminology was the most difficult part of the PE task because MT did not get most words right. More time was needed to look up the words and confirm everything. So, two team members had to help out. That disorganized our work plan a bit. In contrast, Team 2 trainees reported different types of challenges, often related to the AI’s “over-smoothness” or subtle contextual misinterpretations, requiring more nuanced refinement. According to T1T5, the chatbot’s output was surprisingly fluent, but rather a little too generic: I think AI did not get the right tone. That means, even though the translation was fluent, we knew we had significant work to do on the text. It needed careful reading of the ST to capture all the nuances. T1T5’s perception was shared by the group members, who were aware not to over-rely on AI fluency. Another challenge was consistency. T2T3 and T2T4 reinforced this idea, stating, “The biggest challenge was consistency. The AI would translate a term one way in one paragraph and differently later on, even for the same concept,” and “We had to enforce consistency throughout, by ourselves.” In addition to team-based challenges, trainees encountered specific individual challenges. For example, T1T6 and T2T5 experienced technical difficulties with slow-to-respond PCs that slowed their pace; T2T1 had connection problems at one point, though, “Luckily, it didn’t last for a long time,” and T2T2’s computer froze during the task and had to be restarted. According to T2T2, “I would have lost all the data had our team not been working on a cloud-based document.” Team cohesion: The nature of the PE task affected the internal dynamics and sense of unity within each team. Team 1’s challenges with the frequently subpar MT output sometimes cultivated a strong camaraderie, but also resulted in individual frustration. T1T4 noted that the substantial workload generated by the MT brought the team closer together: “We knew we were all in this together, and we had to become like a fighting machine. This spirit brought us even closer together. The goal was to do a good job for the client.” In addition to their sense of togetherness, trainees also complained about MT output. As T1T1 stated: Most of us knew the MT system was not sophisticated and the translation needed improvement. We knew that without human effort, the client would not accept the translation output. For this reason, we had to work together to achieve success. However, despite the overarching sense of togetherness, T1T5 maintained, “I felt like I was constantly battling the MT on my section. Because it was stressful, it sometimes felt like I was working alone on the text, though everyone was talking around me.” In contrast, Team 2’s cohesion was often established through collaborative problem-solving concerning the AI’s output, which necessitated further discussion and collective decision-making on stylistic and contextual improvements. The discussions emphasized the contextual meaning rather than the fluency of the language. According to T2T3, Our discussions were more about how to make the smooth translation of AI truly reflect the meaning of the ST. The English sounded so perfect, but we knew we had to add the right meaning to it. That increased our engagement with the text. Team 2’s cohesion, therefore, was informed by the need to provide context for AI-generated output. That led to increased internal unity in coordinating the different tasks. As T2T3 explained, there was a consistent need to ensure that members charged with highlighting ST words and phrases with nuances provided the input necessary to reinforce the meaning of the TT, stating, We needed to coordinate well with the members charged with underlining areas of the ST that carried the most meaning nuances. We knew that AI left out some meaning, maybe something that only humans could find. That was our job. Communication: While communication remained important, its focus and nature shifted depending on the technology used in the translation. For Team 1, communication primarily centered on identifying errors, developing correction strategies, and ensuring consistency in addressing common MT flaws. As T1T5 noted, Most of our discussions were about specific MT errors. We discussed the different error types, particularly those related to tenses, and found ways to resolve them. Often, we disagreed on how to resolve the problems, but we always ended up with a solution. T1T1 remarked that “we set up a second shared document just for MT error patterns so we simultaneously see each other’s fixes and arrive at conclusions quickly.” The idea of setting up a second shared file specifically for debating and resolving MT errors was commended by all team members. P1T6 claimed, “By isolating the MT errors in a separate document, we could focus on the fixes. I think this helped us tremendously in meeting the deadline.” Unlike Team 1, Team 2’s communication tended to be more interpretive, focusing on refining the AI’s output, discussing stylistic nuances, and aligning on subtle contextual adaptations. For example, T2T2 explained the team had a lengthy discussion on the “feel” of the translation. This was because “AI got most words, but we had to discuss how to make it sound natural for the target audience. We didn’t want the translation to read like AI wrote it.” Other members corroborated the assertion, with T2T5 maintaining, “in addition to ensuring the terminology was correct, we focused on making the conversation more ‘human.’ I think we spent too much time on this aspect because overall, I think AI did a good job.” However, T2T3 felt that the discussion involved more brainstorming than PE, arguing that, Discussions involved more brainstorming than PE. We kept asking questions like “Does this convey the right meaning or tone?” rather than just “Is this grammatically correct?” I think the focus shifted a bit here, and we didn’t exercise our fluency or grammatical competency. Discussing both language and meaning may be more fulfilling, I think. The study highlights specific post-translational challenges stemming from variations in MT and AI outputs. Team 1 encountered difficulties with Baidu’s inaccurate and literal translations, which necessitated extensive adjustments to grammar and terminology, disrupted their workflow, and led to frustration; however, it also promoted camaraderie among team members. Team 2 addressed issues related to Kimi AI’s fluent yet contextually flawed outputs, necessitating subtle adjustments for tone and consistency. Additionally, technical difficulties, such as slow computers, presented personal challenges. Team 1’s communication revolved around correcting errors and utilized a shared document for greater efficiency, while Team 2 prioritized interpretive discussions to improve contextual understanding and style. These interactions influenced team cohesion, as Team 1 bonded over common challenges, and Team 2 came together through joint problem-solving efforts. The findings are consistent with existing research on translation and team dynamics. Team 1 faced difficulties with error-prone MT outputs, necessitating substantial grammatical and terminological corrections. This resonates with the observations of O’Brien (2011) and Guerberof-Arenas (2013), who indicate that low-quality MT leads to increased PE effort and disrupts workflows. In contrast, Team 2 encountered issues with AI-generated outputs that are fluent but contextually inaccurate, aligning with the perspectives of Koponen (2016) and Bowker and Buitrago Ciro (2019). These challenges underscore the need for nuanced adjustments to enhance meaning and consistency. Moreover, Team 1’s camaraderie in stressful situations supports Hackman’s (2002) assertion that shared challenges foster unity, while Team 2’s collaborative problem-solving mirrors the insights of Risku and Dickinson (2017) regarding interpretive teamwork. The differing communication styles are notable, with Team 1 concentrating on error correction (Karamanis et al., 2011; Krings, 2001) and Team 2 focusing on stylistic nuances (LeBlanc, 2013; Olohan, 2021), highlighting the various demands of PE. Integrating quantitative and qualitative findings: Employing a convergent parallel mixed-methods approach, quantitative (time, errors, and readability) and qualitative (stress and communication) data were integrated through a matrix, analyzed separately, and synthesized to triangulate insights. It was observed that Team 2 took longer for PE (60 minutes) compared to Team 1 (48 minutes), reflecting their qualitative emphasis on interpretive discussions (T2T3, T2T2), which was influenced by Kimi AI’s fluent yet contextually inaccurate output (T2T5). This was consistent with their lower error count (12 vs. 19), as stress from TDJs (T2T4) and a focus on meaning over grammar (T2T3) hindered fluent error detection, leading to a reduced performance score (92% vs. 95%). Team 1’s swift error correction (T1P6) and team spirit (T1T4) alleviated TDJ stress (T1T3), allowing for quicker completion and higher performance, despite the flaws in Baidu’s translation (T1T1). 5. Discussion 5.1. Summary of Results This study analyzed the PE of technical texts under TDJs, comparing Team 1 (Baidu MT, MT) with Team 2 (Kimi AI translation) using a convergent parallel mixed-methods approach. The quantitative results indicated that Team 2 required more time to post-edit Kimi AI’s output, which, while slightly more readable, demanded careful adjustments. Team 1 corrected a higher number of errors (19 vs. 12) due to Baidu’s output being more error-prone. However, Team 2 achieved a superior performance score (95% vs. 92%), reflecting Kimi AI’s advantages over Baidu in terms of fluency and accuracy, although both teams faced difficulties with terminology and institutional names. The increased resource consultation by Team 1 appears to correlate with the task’s difficulty, the specific challenges encountered, and the time invested. In contrast, Team 2 depended on the AI tool as a reference. Qualitative data revealed widespread TDJ stress, which affected their focus on quality, with both teams prioritizing speed. Team 1’s effective error correction and teamwork stood in contrast to Team 2’s interpretive dialogues and cooperative problem-solving. Both teams confronted challenges: Team 1 dealt with Baidu’s grammatical inconsistencies, while Team 1 grappled with Kimi’s “over-smoothness” and inconsistent terminology. Although both teams engaged in proactive planning, stress and technical difficulties disrupted their workflows. 5.2. Implications for Translator Training The findings highlight important implications for translator training within the PATT framework, focusing on areas such as TDJ preparation, MTPE, AI integration, and strategies specific to different text types. First, Team 2’s extended 60-minute discussion of Kimi AI’s contextually flawed outputs highlights the necessity of training students in PE to handle AI-generated texts under time constraints. Training programs should replicate AI outputs to help develop skills in enhancing tone, style, and consistency, utilizing collaborative tools (e.g., cloud-based documents) for improved discussions and collaboration. This approach addresses the PATT gap related to AI integration, which aligns with industry trends where AI tools, such as ChatGPT and Kimi AI, are increasingly employed for drafting (Bowker & Buitrago Ciro, 2019). Activities could include timed PE exercises using AI translations of technical texts, allowing trainees to focus on identifying critical contextual errors rather than merely achieving fluency, thereby improving both efficiency and quality. Furthermore, training programs need to strike a balance between AI usage and traditional resources, such as online dictionaries, term banks, glossaries, websites, and corpora, which provide translators, particularly those in teams, with a wider array of options. Second, Team 1’s superior error correction (19 errors) and performance (92%) reveal that low-quality MT necessitates strong error detection skills. Training should incorporate simulations featuring error-prone MT outputs and promote collaboration through tools such as shared error logs (T1P6) during team-debriefing sessions (TDJs). This approach addresses the gaps in MTPE training, equipping students for real-life scenarios where MT errors, such as literal translations, frequently occur (Guerberof-Arenas, 2012; O’Brien, 2011). Engaging in role-based exercises, such as designating terminology specialists (T1T2), can boost team efficiency and ensure preparation for demanding professional situations. Third, the pervasive TDJ stress affecting quality focus requires stress management training. Incorporating techniques such as mindfulness and time prioritization, along with stress measurement tools (e.g., NASA-TLX), can help students manage cognitive load (Sweller, 1988). This could address the PATT gap in TDJ preparation, enabling trainees to balance speed and quality under pressure, as stress diminishes focus on perfection. Simulations should mimic TDJ scenarios, using stress scales to quantify impacts, aligning with Lazarus and Folkman’s (1984) stress appraisal theory. Fourth, the shortcomings in terminology of both tools emphasize the necessity for training tailored to specific text types. Programs should incorporate terminology management with glossaries or databases, especially for technical documents, to bridge the PATT gap related to text-type emphasis. Practical exercises could include developing term bases for industry-specific language to improve the quality of professional translations (Koponen, 2016). Furthermore, the distinct dynamics within teams, such as Team 1’s rapport compared to Team 2’s collaborative approach, indicate that training should promote flexible roles in response to translation outcomes. By simulating a range of MT and AI tasks, trainees can learn to utilize collaborative skills, thus addressing the PATT gap concerning teamwork dynamics in TDJs (Hackman, 2002). By implementing these comprehensive strategies, PATT can enhance employability for graduates and effectively prepare trainees for the evolving needs of the translation field. 6. Conclusion This research examined the process of producing machine- and AI-translated technical documents under tight deadlines to assess their impact on efficiency and team dynamics. Utilizing a convergent parallel mixed-methods approach, 12 trainee translators were split into two groups within a simulated translation agency: Team 1 post-edited Baidu’s MT, while Team 2 worked on output from Kimi AI. The main findings reveal that Team 2 took longer (60 minutes vs. Team 1’s 48 minutes) due to interpretive discussions resulting from Kimi AI’s fluent but contextually flawed text, which was somewhat more readable. Both teams achieved high scores, with Team 2 edging slightly ahead (95% vs. 92%). Team 1 made more corrections, applying effective error correction strategies and team spirit to tackle Baidu’s numerous mistakes, characterized by grammatical errors. Both tools faced difficulties with terminology and institutional names; Team 1 sought additional resources to overcome Baidu’s issues, whereas Team 2 depended significantly on the AI tool. The considerable stress from TDJs, as reported by both teams, led to a focus on speed over quality, with technical issues further hindering their workflows. These results have important implications for translator education within the PATT framework. Training programs need to mimic AI and MT PE in TDJs, guiding trainees in refining AI outputs for tone and consistency while identifying errors in subpar MT through collaborative tools such as shared error logs. Techniques for managing stress, aided by tools such as stress scales, can alleviate pressure related to TDJs, improving focus on quality. Effective terminology management, utilizing glossaries, is crucial for technical documents. Promoting adaptive team roles also enhances collaboration and communication. These strategies tackle PATT shortcomings in TDJ preparation, MTPE, AI integration, and text-specific training, equipping graduates for the demanding and evolving translation industry. 7. Limitations The small sample size of 12 trainee translators in this study limits the ability to generalize findings to larger populations or professional contexts. The use of a single technical news article restricts insights into various text types, such as legal or literary translations, which may lead to missing genre-specific PE challenges. While the 80-minute simulation aims to reflect tight deadlines, it may not accurately represent the ongoing pressure of actual translation workflows, potentially underestimating stress effects. The reliance on self-reported reflections for qualitative data may introduce bias, as trainees could either exaggerate or downplay their experiences. Uncontrolled variables, such as differences in team composition, prior knowledge of Baidu or Kimi AI tools, or individual language skills, could affect results and comparisons between teams. Additionally, focusing on one MT and one AI tool limits the study’s conclusions regarding other tools. Future studies with larger samples, varied texts, and objective measurements could help address these limitations, thus increasing the applicability of PATT training. Declarations Disclosure statement The authors report there are no competing interests to declare . Data availability Data generated during this research is available with the corresponding author upon reasonable request. Funding This work was supported by the Guangdong Province Postgraduate Education Innovation Project under Grant 2023JGXM-066. Ethical Approval This study was conducted under the guidelines of the Helsinki Declaration and with approval from the ethics department of the School of Foreign Languages, Guangdong University of Finance and Economics (Approval No. SFLGDUFE202501211 of Jan. 21, 2025). Informed consent Written informed consent was obtained from participants on January 26, 2025, prior to the study. References Apandi, A., & Afiah, D. S. S. (2019). Project based learning in translation class. Academic Journal Perspective: Education, Language, and Literature , 7 (2), 101-108. https://doi.org/10.33603/perspective.v7i2.2656 Apandi, A., & Afiah, D. S. S. (2019). Project based learning in translation class. Academic Journal Perspective: Education, Language, and Literature , 7 (2), 101-108. Apandi, Apandi, and Devi Siti Sihatul Afiah. "Project based learning in translation class." Academic Journal Perspective: Education, Language, and Literature 7.2 (2019): 101-108. Bowker, L. (2004). What does it take to work in the translation profession in Canada in the 21st century? Exploring a database of job advertisements. Meta, 49 (1), 960–972. Bowker, L., & Buitrago Ciro, J. (2019). Machine translation and global research: Towards improved machine translation literacy in the scholarly community . Emerald Publishing. Bundgaard, K. (2017). Translator attitudes towards translator-computer interaction– Findings from a workplace study. Hermes , (56), 125-144. https://doi.org/10.7146/hjlcb.v0i56.97228 Buysschaert, J., Fernández-Parra, M., Kerremans, K., Koponen, M., & Van Egdom, G. W. (2018). Embracing digital disruption in translator training: Technology immersion in simulated translation bureaus. Tradumàtica , (16), 0125-133. https://doi.org/10.5565/rev/tradumatica.209 Carl, M., Gutermuth, S., & Hansen-Schirra, S. (2015). Post-editing machine translation: A usability test for professional translation settings. In Psycholinguistic and cognitive inquiries into translation and interpreting (pp. 145-174). John Benjamins Publishing Company. Chesterman, A. (2005). Problems with strategies. In K. Károly & Á. Fóris (Eds.), New trends in translation studies: In honour of Kinga Klaudy (pp. 17–28). Akadémiai Kiadó. García González, M., & Veiga Díaz, M. T. (2015). Guided Inquiry and Project-Based Learning in the field of specialised translation: a description of two learning experiences. Perspectives , 23 (1), 107-123. González-Davies, M., & Raído, V. E. (2018). Situated learning in translator and interpreter training: Bridging research and good practice. In Situated Learning in translator and interpreter training (pp. 1-11). Routledge. Gouadec, D. (2007). Translation as a profession . Amsterdam, Netherlands: John Benjamins. Guerberof-Arenas, A. (2013). The role of post-editing in translation workflows: Productivity and quality. The Journal of Specialised Translation, 20 , 76–92. Hackman, J. R. (2002). Leading teams: Setting the stage for great performances . Harvard Business Review Press. Hackman, J. R. (2002). Leading teams: Setting the stage for great performances . Harvard Business Review Press. Hariyanto, S. (2016). Assessment in Translation Research, Teaching, and Industry. In The 2016 International Translation and Interpreting Symposium (pp. 236-244). Karamanis, N., Luz, S., & Doherty, G. (2011). Translation practice in the age of computer-assisted translation: The impact of technology on workflow. The Translator, 17 (2), 231–250. https://doi.org/10.1080/13556509.2011.10799487 Kiraly, D. (2000). A social constructivist approach to translator education: Empowerment from theory to practice . Manchester, England: St. Jerome. Kiraly, D. (2005). Project-based learning: A case for situated translation. Meta , 50 (4), 1098-1111. https://doi.org/10.7202/012063ar Kiraly, Don. "Project-based learning: A case for situated translation." Meta 50.4 (2005): 1098-1111. Konttinen, K., Veivo, O., & Salo, P. (2019). Translation students’ conceptions of translation workflow in a simulated translation company environment. The Interpreter and Translator Trainer , 14 (1), 79–94. https://doi.org/10.1080/1750399X.2019.1619218 Koponen, M. (2010). Assessing machine translation quality with error analysis. Mikael: Kääntämisen ja tulkkauksen tutkimuksen aikakauslehti , 4 . Koponen, M. (2016). Is machine translation post-editing worth the effort? A survey of research into post-editing and effort. The Journal of Specialised Translation, 25 , 131–148. Kozyar, M. M., Pasichnyk, S. M., Kopchak, M. M., Burmakina, N. S., & Suran, T. (2022). Simulation-Based Learning as an Effective Method of Practical Training of Future Translators. Journal of Curriculum and teaching , 11 (1), 298-308. Krings, H. P. (2001). Repairing texts: Empirical investigations of machine translation post-editing processes . Kent State University Press. Künzli, A. (2004). Risk taking: Trainee translators vs. professional translators. A case study. The Journal of Specialised Translation, 2 , 34–49. Kussmaul, P. (1995). Training the translator . Amsterdam, Netherlands: John Benjamins. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping . Springer Publishing Company. LeBlanc, M. (2013). Translators’ perceptions of machine translation: A Canadian survey. The Interpreter and Translator Trainer, 7 (2), 193–214. https://doi.org/10.1080/1750399X.2013.845209 Li, D., Zhang, C., & He, Y. (2015). Project-based learning in teaching translation: students’ perceptions. The Interpreter and Translator Trainer , 9 (1), 1-19. https://doi.org/10.1080/1750399X.2015.1010357 Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A temporally based framework and taxonomy of team processes. Academy of Management Review, 26 (3), 356– 376. https://doi.org/10.5465/amr.2001.4845785 Massey, G., & Ehrensberger-Dow, M. (2017). Machine translation and cognitive load: Insights from eye-tracking and retrospective protocols. The Journal of Specialised Translation, 28 , 138–155. McDermid, C. (2025). Allyship and social annotation software in interpreter education. The Interpreter and Translator Trainer , 19 (2), 137-154. https://doi.org/10.1080/1750399X.2025.2488193 Mitchell-Schuitevoerder, R. (2013, October). A project-based methodology in translator training. In Tracks and Treks in Translation Studies: Selected Papers from the EST Congress, Leuven 2010 (pp. 127-142). John Benjamins Publishing Company. Mo, A., & Man, D. (2017). The ecosystem of translator workstation: Learning electronic tools in a training program for professional translators in China. Babel , 63 (3), 401-422. https://doi.org/10.1075/babel.63.3.06aip Morón, M., & Calvo, E. (2018). Introducing transcreation skills in translator training contexts: A situated project-based approach. The Journal of Specialised Translation , 29 , 126-148. O’Brien, S. (2011). Towards predicting post-editing productivity. Machine Translation, 25 (3), 197–215. https://doi.org/10.1007/s10590-011-9096-2 O’Brien, S. (2021). Post-editing. In Handbook of translation studies (pp. 177-183). John Benjamins Publishing Company. Olohan, M. (2021). Translation and technology: The evolution of translation practice . Routledge. Olvera Lobo, M. D., Castro Prieto, M. R., Quero Gervilla, E., Muñoz Martín, R., Muñoz Raya, E., Murillo Melero, M., ... & Domínguez López, C. (2005). Translator training and modern market demands. Perspectives , 13 (2), 132-142. https://cris.unibo.it/handle/11585/697109 Olvera Lobo, M. D., Robinson, B., Castro Prieto, R. M., Quero Gervilla, E., Muñoz Martín, R., Muñoz Raya, E., ... & Díez Lerma, J. L. (2007). A professional approach to translator training (PATT). Meta , 52 (3), 517-528. https://doi.org/10.7202/016736ar Park, D., & Padó, S. (2024). Multi-Dimensional Machine Translation Evaluation: Model Evaluation and Resource for Korean https://doi.org/10.48550/arXiv.2403.12666 Prieto-Velasco, J. A., & Fuentes-Luque, A. (2018). A collaborative multimodal working environment for the development of instrumental and professional competences of student translators: an innovative teaching experience. In Situated Learning in Translator and Interpreter Training (pp. 76-91). Routledge. Pym, A. (2009). Translator training. In Oxford companion to translation studies (Pre- print). Tarragona, Spain: Intercultural Studies Group. Pym, A. (2009). Translator training. Proyecto para El Oxford companion translation studies . Risku, H., & Dickinson, A. (2017). Translation as teamwork: A situated approach to translators’ collaboration. Translation Spaces, 6 (1), 81–99. https://doi.org/10.1075/ts.6.1.05ris Risku, H., & Dickinson, A. (2017). Translation as teamwork: A situated approach to translators’ collaboration. Translation Spaces, 6 (1), 81–99. https://doi.org/10.1075/ts.6.1.05ris Salas, E., Shuffler, M. L., Thayer, A. L., & Bedwell, W. L. (2015). Understanding and improving teamwork in organizations: A scientifically based practical guide. Human Resource Management, 54 (4), 599–622. https://doi.org/10.1002/hrm.21628 Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12 (2), 257–285. https://doi.org/10.1207/s15516709cog1202_4 Tekwa, K., Su, W., & Li, D. (2024). Web 2.0 technologies and translator training: assessing trainees’ use of instant messaging as a collaborative tool in accomplishing translation tasks. Humanities and Social Sciences Communications , 11 (1), 1-14. https://doi.org/10.1057/s41599-024-02934-5 Tsai, Y. (2025). Tuning language and lyrics: a case study of song translation in the translator training classroom. The Interpreter and Translator Trainer , 1-19. https://doi.org/10.1080/1750399X.2025.2507541 Additional Declarations No competing interests reported. Supplementary Files Appendix1.docx Appendix2.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6892673","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":519189884,"identity":"e3183258-822f-4de4-8e3f-09ee6a0c5abb","order_by":0,"name":"Honghui Hu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCklEQVRIie3RMUvDQBTA8XccmOUkji8U+hkeZBKD/SoXCjfdEPALRApmaZ0r+CF0czwJxCWLW0oc2qUuHTIaEPXq5JI0o9D7D2847scddwAu1z/M5+x609CX8L0UEKL9muwnQTbLQ0zMOJgbS9QAQmWhUDQmpEoOJFBpQqS3+GH1/loLGYHvaYL2qVuwpSZKaBvf1frqQkgFwXxHbFF2E46aJBKPb2utRqcfOZA9l7ObbnJiiRGWpKtSjYT8hskhIkQRp4Ly8KzyCksMEB4g6M1yjqTsI2t+fi+nAstt8rzoIZOcZW3zGdmvfNlUO3k59rPp47rtIX/vSL9zP8wgAOCtB250uVyuY+sHeBNU2a0oRswAAAAASUVORK5CYII=","orcid":"","institution":"Guangdong University Of Finances and Economics","correspondingAuthor":true,"prefix":"","firstName":"Honghui","middleName":"","lastName":"Hu","suffix":""}],"badges":[],"createdAt":"2025-06-14 08:08:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6892673/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6892673/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92160823,"identity":"5c735d8c-bedc-49d8-a664-b32ea6c0fe9f","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":189890,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscriptanonymous.docx","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/5dc11d716cc3a09c5d3f5b3e.docx"},{"id":92162489,"identity":"88ada38d-53e1-4e73-bc06-69c7b5cef81e","added_by":"auto","created_at":"2025-09-25 10:18:53","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3413,"visible":true,"origin":"","legend":"","description":"","filename":"fc7cbe1bd7d543b68f924e219ee7ac0a.json","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/73d39dc122751e48202dd3d1.json"},{"id":92161885,"identity":"08a86801-3038-435f-a75d-a11993cbc43d","added_by":"auto","created_at":"2025-09-25 10:10:53","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33584,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/bc2cc22c2580430a64daf721.docx"},{"id":92161888,"identity":"c621c564-e86a-4f30-8fef-eb1ee1834fec","added_by":"auto","created_at":"2025-09-25 10:10:53","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37353,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix2.docx","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/4cc11c2a4be1e517f86849e4.docx"},{"id":92160830,"identity":"22a40145-56e6-4deb-b1af-6803942662e0","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110500,"visible":true,"origin":"","legend":"","description":"","filename":"fc7cbe1bd7d543b68f924e219ee7ac0a1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/be4569ffbdd66b2b19818277.xml"},{"id":92160818,"identity":"5fda4fd4-811f-4771-936e-ad291868235d","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":74176,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/4e9316868be14fc6a8f34897.png"},{"id":92160826,"identity":"5e552c92-b766-417d-a1cb-f7e6691cae30","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":234797,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/92692b4c8942de84eedd5cb0.jpeg"},{"id":92160828,"identity":"4df1df32-4b01-4ba7-ad74-de9ad2575798","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":313520,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/7e07ea1df3c7802c3f07a6a0.jpeg"},{"id":92162491,"identity":"63b1d957-6292-4bd2-a9a5-e91b1f74d44c","added_by":"auto","created_at":"2025-09-25 10:18:53","extension":"emf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114692,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.emf","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/feec236150207a27f3341b98.emf"},{"id":92160822,"identity":"feadd289-5ec6-47f8-8c2e-70ac1e6601e4","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28379,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/95215f2bf512e00d5d0ff75a.png"},{"id":92161890,"identity":"41998621-b8c7-42d8-bde8-d93cb536698c","added_by":"auto","created_at":"2025-09-25 10:10:53","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41378,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/44334774186b80bbae525db3.png"},{"id":92160832,"identity":"b3d272d1-5578-4fc6-a4f2-28a91e694d94","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":140383,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/5c952eddfdb2a695debc76ef.png"},{"id":92160831,"identity":"9c7dbfdb-ffb6-475b-a615-522b0bb2d460","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":149888,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/4f9673ce7a4116560b63c25a.png"},{"id":92161891,"identity":"59086bf8-9fe8-446c-8384-ef42e432bce9","added_by":"auto","created_at":"2025-09-25 10:10:53","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108990,"visible":true,"origin":"","legend":"","description":"","filename":"fc7cbe1bd7d543b68f924e219ee7ac0a1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/b355074289bbb2d0d2dcefa9.xml"},{"id":92160834,"identity":"a57f6e79-d6b1-4c42-953c-e8e41464583a","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"html","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118160,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/3931d53f98cd165a16f94ee7.html"},{"id":92160824,"identity":"d70069b4-552c-4b55-95b3-4a2f5087127a","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":58433,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTranslation process for Teams 1 and 2\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/8f436b28f4578bd5248d6f2c.png"},{"id":98434061,"identity":"d0feb679-d51e-4819-ac22-0e4192d69722","added_by":"auto","created_at":"2025-12-17 16:51:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1008190,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/66deceaf-65a1-48a3-8e0d-b1b6a3f9fffb.pdf"},{"id":92160817,"identity":"83f73af9-3807-49d3-b2df-963a997284da","added_by":"auto","created_at":"2025-09-25 10:02:53","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":33584,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/e10e782eb349c268cbf431c0.docx"},{"id":92162490,"identity":"0bbbc093-8a46-4ad9-ae7e-68afea74e191","added_by":"auto","created_at":"2025-09-25 10:18:53","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":37353,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix2.docx","url":"https://assets-eu.researchsquare.com/files/rs-6892673/v1/3af429e581d40e4bbf549a16.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Team dynamics and efficiency in post-editing: Machine translation systems versus artificial intelligence chatbots under tight deadline conditions","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe translation industry is undergoing a significant transformation influenced by artificial intelligence (AI) and automation, necessitating training that meets market demands and professional standards. Rapid developments in machine translation (MT) and AI chatbots have transformed workflows, making the post-editing (PE) of AI- and MT-generated texts a core aspect of professional practice (Bowker \u0026amp; Buitrago Ciro, 2019). Tight deadline jobs (TDJs) are becoming more prevalent, necessitating efficiency, collaboration, and the ability to adapt under pressure. However, conventional translator training often falls short, emphasizing linguistic accuracy over the practical, technology-based skills essential in global markets (Pym, 2009). This gap highlights the urgent need for curricula that replicate real-world conditions, incorporating AI tools, machine-translated post-editing (MTPE), and collaborative processes to equip graduates with the skills demanded by industry.\u003c/p\u003e\n\u003cp\u003eA professional approach to translator training (PATT) meets this need by simulating professional workflows through project-based learning (PBL) and workplace scenarios (Olvera Lobo et al., 2007). It prioritizes learner-centered education and promotes translation competence through roles such as translator, terminologist, and project manager, utilizing tools such as Translation Memory systems and collaborative platforms (e.g., Aula.int, Basic Support for Cooperative Work [BSCW]). However, PATT encounters challenges, including a limited emphasis on TDJs, inadequate integration of MTPE, and a lack of exploration into AI chatbots or strategies specific to certain text types (Konttinen et al., 2019; Morón \u0026amp; Calvo, 2018). These limitations hinder its ability to adequately prepare students for high-pressure, technology-driven settings, underscoring the need for research that bridges academic training with industry requirements.\u003c/p\u003e\n\u003cp\u003eThis study examines the impact of PE machine- and AI-translated technical texts in TDJ conditions, with a focus on process efficiency and team dynamics. It addresses two research questions: (1) What impact do machine- and AI-translated texts have on the PE process in a time-sensitive professional environment? (2) How do TDJ conditions affect collaborative team interactions? Conducted in a simulated translation training setting with 12 trainee translators divided into two teams, one PE a Baidu MT output and the other a Kimi AI output, the study simulates a real-world scenario involving a technical news article. The design meets industry requirements for swift and precise translations.\u003c/p\u003e\n\u003cp\u003eKey terms in this study consist of the following: team cohesion, which refers to how unified and supported trainees felt, reflecting their collaboration and team spirit; challenges encountered, representing obstacles such as technical difficulties or time limitations; communication, which encompasses the exchange of ideas, including both initial planning and ongoing discussions; TDJ perceptions, indicating trainees’ experiences of stress and pressure within tight timeframes; and team organization, highlighting the arrangement of roles and tasks to effectively manage the project.\u003c/p\u003e\n\u003cp\u003eThis study enhances the PATT framework by tackling essential gaps in TDJ preparation, MTPE, AI integration, and training tailored to specific text types. Exploring the connections between technology and teamwork in high-pressure situations reveals how translator training can be enhanced to effectively leverage contemporary workflows. The results influence curriculum development by promoting stress management techniques, collaborative tools, and terminology strategies, all aimed at boosting employability. Furthermore, this research advances translation pedagogy by providing empirical support for AI and MT PE practices, aligning with market demands and equipping graduates for the evolving, AI-centered translation industry (Pym, 2009; Olvera Lobo et al., 2007).\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003eThe literature review section focuses on professional translation training.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1. Professional Approach to Translator Training\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePATT is a cutting-edge approach to translation instruction that emulates real-world professional environments by assigning students specific roles in a PBL setting (Olvera Lobo et al., 2007). This method incorporates information and communication technology to simulate a translation agency. In a PATT instructional environment, trainees assume various roles within translation teams, engaging in tasks that typically involve documentation, terminology management, translation, revision, and formatting. These activities develop essential skills in teleworking, teamwork, and interdisciplinary collaboration (Alvera Lobo et al., 2007). The process reflects a professional translation workflow, encompassing pre-translation (research and terminology), information storage, linguistic conversion (representation), and delivery of the final text (communication) (Olvera Lobo et al., 2005). PATT prioritizes learner-centered education, resonating with the Bologna Declaration’s emphasis on career-oriented teaching and promoting student autonomy (Olvera Lobo et al., 2007). This approach enhances translation competence by integrating communicative, extra-linguistic, instrumental-professional, and strategic sub-competences, thereby embedding practical, market-relevant skills within the curriculum.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA significant contribution to PATT was made at the University of Granada, Spain (Olvera Lobo et al., 2007) as part of its four-year Translation and Interpreting degree program. This was implemented using the Aula.int platform, enhanced by the BSCW system. This setup enabled asynchronous collaboration, allowing students to organize tasks, share documents, and communicate online, effectively preparing them for the teleworking aspects of the translation profession. The approach provided numerous advantages by simulating a translation agency through the Aula.int platform. It immerses students in a realistic professional setting, assigning roles such as translator, terminologist, and project manager, thereby fostering practical skills in documentation, terminology, and formatting (Olvera Lobo et al., 2007). PATT enhances both teleworking and teamwork via the BSCW platform, which facilitates asynchronous collaboration and prepares students for the global market. It combines interdisciplinary skills from various course modules, encouraging a comprehensive understanding of translation competence, which includes communicative and instrumental-professional sub-competences (PACTE, 2000). By promoting student autonomy and self-directed learning, PATT increases engagement and motivation while addressing technical challenges, effectively equipping students for the demands of the translation profession.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2. Related Studies on PATT\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudies employing a professional translation approach have been theoretical, practical, or a combination of both. Pym (2009) is one of the most theoretically focused scholars advocating for training that meets job demands, particularly in the digital era. In his 2009 work, he critiqued traditional translator training for its instructional, teacher-centered nature, where students translate primarily for assessment against literal models by often non-professional teachers, which fails to develop professional competencies. He contended that this transmissionist model overlooked market requirements, leading to graduates who are ill-prepared for industry expectations (Bowker, 2004; Gouadec, 2007). Pym urged that training should reflect real-world practice through authentic tasks, portfolios, and social constructivist approaches, enabling students to collaboratively build skills such as problem-solving and client engagement (Kiraly, 2000). He also emphasized the importance of incorporating translation memories and specialized software (e.g., localization tools) to align with industry standards and specific sectors, highlighting professional strategies that correspond to actual competencies (Künzli, 2004; Kussmaul, 1995).\u003c/p\u003e\n\u003cp\u003eBeyond theoretical frameworks, numerous empirical studies have explored the alignment of training with practice through various methods. In the PBL framework, researchers have emphasized the importance of student-centered, constructivist strategies in translator education, focusing on authentic, collaborative, and technology-oriented tasks. Mitchell-Schuitevoerder (2013) examined the role of reflective group work and blogs in fostering critical thinking within a translation technology module. Morón and Calvo (2018) incorporated transcreation through simulated marketing projects to address intercultural issues, while Li et al. (2015) employed PBL in business translation, implementing real-world projects to enhance practical skills. García González and Veiga Díaz (2015) adopted Guided Inquiry and peer reviews in specialized translation classes to align with higher education objectives. Tekwa et al. (2024) integrated instant messaging into collaborative tasks within a project-based translation and localization training context, while Apandi and Afiah (2019) centered on cultural translation, promoting independent learning.\u003c/p\u003e\n\u003cp\u003eResearch indicates that PBL promotes professional skills, employability, and critical thinking through authentic tasks, such as transcreation and localization (Morón \u0026amp; Calvo, 2018; Tekwa et al., 2024). It encourages collaboration and engagement, evidenced by positive student feedback (Li et al., 2015; Mitchell-Schuitevoerder, 2013). However, some drawbacks include limited preparation time, difficulties in group adaptation, and logistical challenges, such as accessing observation sites (Apandi \u0026amp; Afiah, 2019). The translation of cultural content and the integration of new tools, such as transcreation or instant messaging, present complexities and often lack established frameworks or quantitative validation (Morón \u0026amp; Calvo, 2018; Li et al., 2015). These challenges underscore the need for enhanced resources and research to improve the effectiveness of PBL in translator training.\u003c/p\u003e\n\u003cp\u003eAlongside PBL, PATT researchers have implemented a workplace simulation approach, focusing primarily on workflows at translation companies, authentic project-based tasks, and translator workstations (Konttinen et al., 2019; Mo \u0026amp; Man, 2017; Prieto-Velasco \u0026amp; Fuentes-Luque, 2018). This approach incorporates real-world conditions, MT-assisted translation memory (TM) tasks, and role-based scenarios, such as those involving translation specialists or project managers (Bundgaard, 2017; Kozyar et al., 2022). The tools employed include TM systems, MT, Web 2.0 collaborative platforms, terminology databases, and project management software (Buysschaert et al., 2018). Some studies emphasize the holistic integration of various technologies, while others focus on standard computer-assisted translation (CAT) tools and online resources, which enable students to engage with industry-standard workflows and authentic translation assignments in simulated environments (González-Davies \u0026amp; Raído, 2018).\u003c/p\u003e\n\u003cp\u003eThe simulations have aimed at developing professional translation competence, foreign language proficiency, and technology-based skills, such as using CAT tools and Web 2.0 platforms (Mo \u0026amp; Man, 2017; Prieto-Velasco \u0026amp; Fuentes-Luque, 2018). Organizational, interpersonal, and problem-solving skills have also been fostered, alongside confidence and autonomy (Konttinen et al., 2019; Kozyar et al., 2022). Benefits include enhanced employability through industry-aligned training, improved academic performance, and positive attitudes toward collaborative tools (Buysschaert et al., 2018). Simulations bridge the gap between academic training and professional practice, preparing students for real-world pressures and enabling them to adapt to technological imperfections, ultimately enhancing their ability to think and act like professional translators (Bundgaard, 2017; González-Davies \u0026amp; Raído, 2018).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3. Existing Lacunae and Research Questions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResearch on PBL and simulated workplace environments in the PATT framework uncovers several significant gaps that hinder their relevance to current translation practices. First, both PBL and simulation research largely neglect TDJs, which are prevalent in professional environments and demand swift decision-making and efficiency under pressure. Although PBL emphasizes collaboration (e.g., Mitchell-Schuitevoerder, 2013; Tekwa et al., 2024), it overlooks how collaborative dynamics operate under the time constraints of TDJs, creating a gap in preparing students for high-pressure environments. Furthermore, neither PBL nor simulations under PATT sufficiently investigate MTPE, despite its prominence in contemporary translation workflows (Bundgaard, 2017 mentions MT-assisted TM but does not specifically address PE). The role of AI, especially AI chatbots, is also missing, despite their increasing involvement in translation tasks for terminology assistance and drafting. Finally, the importance of text types, which are essential for customizing translation strategies, is not a focal point in PATT research, despite workplace simulations encompassing diverse tasks (e.g., transcreation, localization) (Konttinen et al., 2019; Morón \u0026amp; Calvo, 2018;). These gaps highlight the need for PATT to incorporate TDJs, MTPE, AI tools, and text-type-specific strategies to better reflect the evolving demands of the translation industry and enhance graduate employability.\u003c/p\u003e\n\u003cp\u003eThis study addresses these gaps by examining the PE of a technical text under strict deadline conditions. It replicates a work environment with two trainee groups: one that post-edits a machine-translated text and another that post-edits an AI-translated text. The research addresses the following research questions:\u0026nbsp;\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eHow do machine- and AI-translated texts affect the PE process under tight deadline conditions?\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eHow do TDJ conditions impact team performance dynamics?\u0026nbsp;\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThis study significantly enhances the PATT framework by tackling major gaps in current PBL and simulated workplace studies, especially concerning TDJs, MTPE, AI integration, and team dynamics.\u003c/p\u003e"},{"header":"3. Methodology","content":"\u003cp\u003eThe approach utilized involved simulating actual practices in the translation industry. The client, a news media company, supplied an 836-character text that needed to be posted online within two hours, requesting an urgent translation. A group of 12 trainee translators was created through simple random sampling and divided into two teams of six. Each team translated the source text (ST) using different tools. Team 1 employed MT systems for their translation, which they subsequently post-edited. In contrast, Team 6 utilized AI chatbots for the same task, followed by PE. This methodology aimed to replicate real-life scenarios faced by translators who sometimes need to deliver translations on tight deadlines, explore the effects of MTPE and AI chatbot PE (AIPE) processes, evaluate teamwork dynamics under these conditions, and collect students' perceptions of various translation tools.\u003c/p\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Participants\u003c/h2\u003e\u003cp\u003eParticipants included 12 trainee translators in their second semester of the MTI program at a Chinese university. This sample aligns with current research in the field (McDermid, 2025; Tsai, 2025), as graduate class sizes are typically small. All participants were enrolled full-time in the MTPE course, where they developed their translation and PE skills, collaborative teamwork, and understanding of the role of AI in facilitating translation. The class consisted of eight female and four male students who were admitted to the program after a careful selection process and an entrance examination that evaluated their translation and writing skills in both Chinese and English. In the MTPE course, they translated texts from Chinese to English in the CAT laboratory, a multimedia classroom specifically designed for translation instruction. Baseline scores were measured based on their performance in prior MTPE and AI-based MTPE activities that preceded this intervention. All ethical requirements were satisfied, and informed consent was obtained from the trainees.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.2. Design\u003c/h2\u003e\u003cp\u003eThe experiment was conducted in Week 14 of the 18-week semester and spanned 80 minutes. From Weeks 3 to 10, students focused on PE texts with different MT systems, including DeepL, Google Translate, Baidu Translate, and Yandex. In Weeks 11 and 12, trainees learned about AI prompting and how to utilize AI chatbots for PE. Specifically, they employed an AI chatbot to translate the text and then, using various prompts, conducted terminology research and enhanced the grammar of English sentence structures. In Week 13, both PE techniques (informed by MT and AI) were reviewed in class to ensure all students had a consistent baseline. They completed three assessments before the intervention\u0026mdash;Week 4 and Week 10 for MTPE, and Week 12 for AIPE. The assessment in Week 10 was a group-based activity. No statistical differences were observed among the trainees.\u003c/p\u003e\u003cp\u003eMoreover, teamwork was a key component of most in-class activities. Students routinely collaborated in teams, learning to allocate tasks, discuss ideas, and adopt various roles within translation scenarios. This pedagogical strategy aligns with PBL (Apandi \u0026amp; Afiah, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), a recognized method for simulating workplace environments that helps students gain insights into both the complex nature of the learning process and the translation process itself (Kiraly, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe intervention took place in the CAT laboratory, where the two teams were positioned at opposite ends, separated by three workstations. This arrangement ensured that discussions held by one team did not interfere with those of the other. Trainees received the text and were informed that the client expected the translated version within 70 minutes. Subsequently, they were provided with instructions on handling the text, as detailed in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Two trained assistants oversaw the experiment using a checklist to maintain consistency.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eTranslated text\u003c/strong\u003e\u003cp\u003eTrainees translated a technical news piece about U.S. export regulations impacting the electronic design automation sector, aligning with the changing requirements of the translation field. This genre exemplifies the high-stakes, urgent nature of technical translation in the global tech industry, while also filling a gap in PATT research regarding TDJs and specific text-type strategies (Mor\u0026oacute;n \u0026amp; Calvo, 2018). The intricate terminology and cross-disciplinary themes (technology, finance, geopolitics) made it suitable for assessing machine- and AI-translated text PE under pressure, and fostered the exploration of team dynamics. By focusing on a text that reflects real-world translation challenges, this study can enhance the employability of trainees within the PATT framework.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eTranslation assessment\u003c/strong\u003e\u003cp\u003eThe translation assessment methodology draws from the Multidimensional Quality Metrics (MQM), Post-Editing of Machine Translation (PEMT), and the LISA QA Model, emphasizing fluency, accuracy, and terminology as supported by Koponen (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), Lommel \u0026amp; Burchardt (2014), and Park \u0026amp; Pad\u0026oacute; (2024) for MQM; O\u0026rsquo;Brien (2021) and Carl et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) for PEMT; and Hariyanto (2016) and Lommel (2018) for the LISA QA Model. The evaluation criteria encompassed fluency (spelling, collocation/cohesion, word form, punctuation, and stylistic issues such as subject-verb agreement, tense, missing articles, subject-verb complement, and adverb placement) and accuracy (omission, addition, incorrect term, untranslated word/expression, over-translation, under-translation, British vs. American English, and mistranslation). All errors were weighted equally to ensure uniformity and consistency throughout the evaluation process. Three assessors, collectively having over 50 years of experience in translator training, performed the assessment. They initially assessed the translation output independently before engaging in a collaborative discussion to align their scores, thereby ensuring a consistent and standardized evaluation.\u003c/p\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.3. Data Collection and Instruments\u003c/h2\u003e\u003cp\u003eThe translation assessment\u0026rsquo;s data collection process utilized various tools to thoroughly evaluate the PE task. Students used a digital timer to record the start and end times, measuring time efficiency. The translation output was examined for accuracy and fluency, with students noting the resources they referred to, including dictionaries, websites, glossaries, and term banks. Furthermore, students submitted typed personal reflection notes discussing team cohesion, challenges, communication, their perceptions of the translation and PE processes, and team organization. Quantitative data, such as time efficiency metrics, were analyzed with the Excel Data Analysis tool, while qualitative data from reflection notes were assessed using NVivo to gauge their relevance in simulating a real-world workplace environment.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.4. Data Analysis\u003c/h2\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003e3.4.1. Quantitative Data Analysis\u003c/h2\u003e\u003cp\u003eThe differences in scores between the two teams were determined through the evaluation of the PE text. Assessments were made regarding both accuracy and fluency. Three key factors were taken into account: the number of correctly edited errors, unedited errors, and incorrect edits. The total score of 100 was adjusted by subtracting the counts of unedited and incorrect edits to derive the performance score for each team. Fluency errors were assigned a weight of 1, while adequacy errors, deemed more significant for meaning, received a weight of 2. The simplification of error weights aimed to improve scoring consistency. At times, a single error within one sentence could be interpreted in various ways, and in such instances, \u0026ldquo;the analysis with the lowest total error count was selected\u0026rdquo; (Koponen, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, p. 6).\u003c/p\u003e\u003cp\u003eQualitative data: A deductive thematic analysis was conducted to assess the qualitative data obtained from trainees\u0026rsquo; reflections. Identified themes included team cohesion, challenges faced, communication, particularly the discussions before starting the task, perceptions of TDJ, and team organization. The analysis utilized NVivo. Two trained coders meticulously reviewed the data and initially coded it using a predetermined set of codes (Appendix 1). After initial coding and discussion, the coders\u0026rsquo; reliability was determined to be significantly high (Cohen\u0026rsquo;s kappa [k]\u0026thinsp;=\u0026thinsp;0.87). Within the NVivo tool, the codes were examined in the Nodes section, and patterns, similarities, and connections were investigated by organizing them into parent\u0026ndash;child relationships. This approach facilitated the extraction of insights on how the two teams operated, as well as their perceptions of the TDJ and the work environment simulation.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"4. Findings","content":"\u003cp\u003e\u003cstrong\u003e4.1. Quantitative Findings\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e4.1.1. Readability Scores\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKimi\u0026apos;s translation had a Flesch Reading Ease score of 42, making it slightly easier to read than Baidu\u0026apos;s score of 39.3. However, both are categorized as \u0026ldquo;difficult\u0026rdquo; and are suitable for educated readers. In terms of complexity, Baidu\u0026rsquo;s translation recorded a Flesch\u0026ndash;Kincaid grade level of 13.3 and a Gunning Fog index of 15.5, in contrast to Kimi\u0026rsquo;s scores of 12.7 and 15.2, respectively. Kimi\u0026rsquo;s version was lengthier, comprising 574 words and 27 sentences, whereas Baidu\u0026rsquo;s contained 444 words and 20 sentences, despite Baidu\u0026rsquo;s sentences being slightly longer on average (22.20 vs. 21.26 words). Both translations featured 12 paragraphs, and their percentage of complex words was similar (Kimi: 19.16%, Baidu: 19.4%). However, Kimi used a greater number of complex words in total, attributable to its higher word count, as illustrated in Table 1.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e Differences between the Baidu and Kimi translations\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"554\" height=\"214\" src=\"data:image/emf;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e4.1.2. Error Types\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur error type analysis indicated that Kimi AI outperformed Baidu Translate in terms of both fluency and accuracy. In terms of fluency, Kimi AI demonstrated better grammatical consistency, including correct tense and subject\u0026ndash;verb agreement, as well as more natural phrasing, avoiding awkward calques such as Baidu\u0026rsquo;s \u0026ldquo;platformization.\u0026rdquo; However, both tools had issues with adverb placement and missing articles. Regarding accuracy, Kimi AI made fewer critical mistakes, notably avoiding additions (e.g., the term \u0026ldquo;potential\u0026rdquo; in \u0026ldquo;US supply cut\u0026rdquo;) and over-translations; however, it shared some weaknesses with Baidu concerning terminology (e.g., \u0026ldquo;Gai Lun\u0026rdquo; vs. \u0026ldquo;Grain-Physics Electronics\u0026rdquo;) and untranslated institutional names (e.g., BIS). Baidu\u0026rsquo;s literal translations (e.g., \u0026ldquo;market has voted\u0026rdquo;) and omissions (e.g., leaving out the 2025 revenue clause) significantly affected clarity. In conclusion, while Kimi AI produced more idiomatic English, both translations still needed human review for technical terms and nuanced expressions to meet professional-grade quality.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e4.1.3. Time and Resources Consulted\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRegarding the time spent on the task, Team 2 took 60 minutes to post-edit the text, while Team 1 took 48 minutes, which is 12 minutes less. This means Team 1 had more time to proofread and revise the post-edited text before submission. Meanwhile, Team 1 consulted significantly more resources than Team 2 during the PE task. As outlined in Table 2, Team 1 referred to eight resources, including online dictionaries, glossaries, term banks, corpora, social media, and websites. In contrast, Team 2 relied primarily on websites while significantly depending on the AI chatbot for terminological, grammatical, and stylistic references.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Resources consulted by the two teams\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"546\" height=\"229\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e4.1.4. Performance\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBoth teams encountered various fluency and accuracy errors that needed correction. Team 1, which worked on the Baidu MT text, discovered and addressed 19 errors, whereas Team 2, which focused on the AI-translated text, found and corrected 12 errors. The examples presented in Table 3 (refer to Appendix 2 for the complete list of errors) show that although some errors were the same, many others stemmed from the translation tool. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Examples of errors found and fixed by teams\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"497\" height=\"381\" src=\"data:image/emf;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n\u003cp\u003eSince neither team post-edited the same text, we calculated the number of unidentified errors and subtracted that from a total of 100 points. Based on this calculation, Team 1 scored 92% and Team 2 scored 95%. The errors included fluency issues, such as incorrect use of passive voice, punctuation mistakes, and problems with sentence structure.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2. Qualitative Findings\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData analysis highlighted both similarities and differences between the teams, facilitating a better understanding. Perceptual similarities spanned the four themes. While all groups shared common perceptions of TDJs, similarities in team cohesion, organization, and communication were noted before the text translation was conducted using the Baidu Translate MT system and the Kimi AI chatbot.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e4.2.1. Perceptual Similarities\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePervasive nature of the TDJ:\u0026nbsp;\u003c/strong\u003eThroughout both teams, the looming pressure of the strict deadline was a prominent factor, rated between 9 and 10 on a scale of 1\u0026ndash;10. Both teams experienced significant stress and pressure due to the quick turnaround demands. This tight deadline infiltrated their workflow from the beginning, impacting their quality standards, decision-making processes, and overall well-being. The qualitative data suggested it was an underlying factor with a noticeable impact. Selections from the trainees\u0026rsquo; reflection notes support our analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExcerpts from trainees:\u003c/strong\u003e Team 1 Trainee 3 (T1T3) maintained that \u0026ldquo;the timer was like an enemy to us. Everyone minute was precious, and because the time kept ticking, we felt anxious. I was sure we would finish the job, but I doubted the quality of the output.\u0026rdquo; Additionally, T1T1 maintained:\u003c/p\u003e\n\u003cp\u003eFrom the time we received the text, it was clear this would be a scramble. The tension was in the air, and we just wanted to start immediately. Because we wanted to save every second, I believe our first discussion was not well coordinated.\u003c/p\u003e\n\u003cp\u003eMeanwhile, T1T4 expressed how the urgent deadline influenced their thinking, saying,\u003c/p\u003e\n\u003cp\u003eI don\u0026rsquo;t think my mind was on doing a perfect job most of the time. I was rather focused on how we could finish the task within the time limit. I think this thought shifted my focus a bit and influenced my concentration during the PE process.\u003c/p\u003e\n\u003cp\u003eT1T6 corroborated this perception, stating that the anxiety of not finishing in time and not producing the quality the client expected were consistent throughout the entire activity. \u0026ldquo;It felt like an examination that I had to finish in time and also obtain a high score.\u0026rdquo;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTeam 2 trainees had similar perceptions. T2T4 remarked,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTranslating the text was like a sprint. I was always looking at the time, thinking \u0026ldquo;Can we really polish this enough.\u0026rdquo; The pressure, for me, was to do a good translation; I don\u0026rsquo;t think I was thinking about a perfect translation.\u003c/p\u003e\n\u003cp\u003eT2T1 added that they spent a significant amount of time during the initial discussion talking about the tight deadline; \u0026ldquo;luckily, our team leader asked us to try and focus on the task, which, was still a little hard to do.\u0026rdquo; Meanwhile, T2T5 explained that right up until the end, the worry of not producing a good-quality text persisted in their mind:\u003c/p\u003e\n\u003cp\u003eThe translation of Kimi AI was, to be honest, quite good. That gave us a good start, I think. But, even when we submitted the text, I was still worried we did not have enough time to produce a really good quality translation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInitial communication and team organization:\u0026nbsp;\u003c/strong\u003eBefore translating and ultimately subjecting the texts to PE, both teams recognized the critical need for upfront planning, clear role allocation, and effective communication to manage the tight deadline. Their reflections revealed similar, proactive strategies in this crucial initial phase, which established a foundation for their work. T1T2 maintained:\u003c/p\u003e\n\u003cp\u003eWe had a quick huddle to divide the work. I think, as we have done in class several times, we needed a solid plan to avoid getting lost in the chaos of the deadline. So, I was responsible for the terminology work, since I can look up the dictionary faster and I type faster than most of my group members.\u003c/p\u003e\n\u003cp\u003eMost Team 1 and 2 members shared the same ideas, focusing on how the work was divided based on individual team member qualities and skills. For example, T2T6 was responsible for terminology research because they could type quickly and mine information more efficiently: \u0026ldquo;I knew they would assign me to look up words and quickly update the target text. I am better at this than most of my team members.\u0026rdquo;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTwo types of organization were highlighted: One based on the individual skills of members and the other on sharing parts of the text among team members. As T1T5 maintained, \u0026ldquo;Our first ten minutes were all about who does what. We split the text into manageable chunks and assigned them. There was no time for indecision.\u0026rdquo; T2T3 confirmed this organization, noting, \u0026ldquo;Our first step was to assign sections and set internal mini-deadlines. We knew that without a tight organizational structure, the TDJ would overwhelm us. Communication at this stage was vital for alignment.\u0026rdquo; At this stage, communication was brief for both teams, essentially during the first ten minutes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe findings underscore the widespread stress caused by TDJ on two translation teams, affecting their workflow, decision-making, and overall well-being. Trainees expressed anxiety about juggling speed and quality, frequently prioritizing completion over perfection. Effective planning and skill-based role assignment helped alleviate the pressure. These results are consistent with Lazarus and Folkman\u0026rsquo;s (1984) stress research, which indicates that time pressure diminishes cognitive focus, as well as with Chesterman\u0026rsquo;s (2005) observations regarding translators\u0026rsquo; focus on efficiency. Cognitive load theory (Sweller, 1988) elucidates the trainees\u0026rsquo; compromises in quality, while Hackman\u0026rsquo;s (2002) research on team dynamics validates their successful role allocations. Initial communications reflect the findings of Marks et al. (2001) on the importance of concise coordination in high-pressure environments, highlighting common challenges faced in time-sensitive tasks.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e4.2.2. Perceptual Differences\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDivergent experiences\u0026mdash;post-translation dynamics:\u0026nbsp;\u003c/strong\u003eOnce the initial machine-generated or AI-generated translations were available, the unique characteristics of these outputs resulted in significant differences in the challenges faced, team cohesion, and communication patterns that followed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eChallenges encountered:\u003c/strong\u003e The error types and the required PE effort varied significantly, leading to distinct challenges for each team. Team 1 trainees frequently highlighted issues stemming from the raw, often literal, and grammatically inconsistent MT output. For example, T1T1 maintained that \u0026ldquo;The machine-translated text was raw. I think Baidu did not get most words right, so we spent a lot of time fixing grammatical errors and awkward phrasing. To me, this felt like intensive editing rather than translation.\u0026rdquo; Four members of Team 1 shared this perception. T1T6 added that \u0026ldquo;Sometimes the MT completely missed the point or used the wrong terminology. That slowed us down because we had to assign more people to do terminology work rather than post-edit.\u0026rdquo; Meanwhile, T1T2, initially assigned with the terminology task, remarked,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eI think terminology was the most difficult part of the PE task because MT did not get most words right. More time was needed to look up the words and confirm everything. So, two team members had to help out. That disorganized our work plan a bit. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn contrast, Team 2 trainees reported different types of challenges, often related to the AI\u0026rsquo;s \u0026ldquo;over-smoothness\u0026rdquo; or subtle contextual misinterpretations, requiring more nuanced refinement. According to T1T5, the chatbot\u0026rsquo;s output was surprisingly fluent, but rather a little too generic:\u003c/p\u003e\n\u003cp\u003eI think AI did not get the right tone. That means, even though the translation was fluent, we knew we had significant work to do on the text. It needed careful reading of the ST to capture all the nuances.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eT1T5\u0026rsquo;s perception was shared by the group members, who were aware not to over-rely on AI fluency. Another challenge was consistency. T2T3 and T2T4 reinforced this idea, stating, \u0026ldquo;The biggest challenge was consistency. The AI would translate a term one way in one paragraph and differently later on, even for the same concept,\u0026rdquo; and \u0026ldquo;We had to enforce consistency throughout, by ourselves.\u0026rdquo;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn addition to team-based challenges, trainees encountered specific individual challenges. For example, T1T6 and T2T5 experienced technical difficulties with slow-to-respond PCs that slowed their pace; T2T1 had connection problems at one point, though, \u0026ldquo;Luckily, it didn\u0026rsquo;t last for a long time,\u0026rdquo; and T2T2\u0026rsquo;s computer froze during the task and had to be restarted. According to T2T2, \u0026ldquo;I would have lost all the data had our team not been working on a cloud-based document.\u0026rdquo;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTeam cohesion:\u003c/strong\u003e The nature of the PE task affected the internal dynamics and sense of unity within each team. Team 1\u0026rsquo;s challenges with the frequently subpar MT output sometimes cultivated a strong camaraderie, but also resulted in individual frustration. T1T4 noted that the substantial workload generated by the MT brought the team closer together: \u0026ldquo;We knew we were all in this together, and we had to become like a fighting machine. This spirit brought us even closer together. The goal was to do a good job for the client.\u0026rdquo; In addition to their sense of togetherness, trainees also complained about MT output. As T1T1 stated:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMost of us knew the MT system was not sophisticated and the translation needed improvement. We knew that without human effort, the client would not accept the translation output. For this reason, we had to work together to achieve success.\u003c/p\u003e\n\u003cp\u003eHowever, despite the overarching sense of togetherness, T1T5 maintained, \u0026ldquo;I felt like I was constantly battling the MT on my section. Because it was stressful, it sometimes felt like I was working alone on the text, though everyone was talking around me.\u0026rdquo;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn contrast, Team 2\u0026rsquo;s cohesion was often established through collaborative problem-solving concerning the AI\u0026rsquo;s output, which necessitated further discussion and collective decision-making on stylistic and contextual improvements. The discussions emphasized the contextual meaning rather than the fluency of the language. According to T2T3,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur discussions were more about how to make the smooth translation of AI truly reflect the meaning of the ST. The English sounded so perfect, but we knew we had to add the right meaning to it. That increased our engagement with the text.\u003c/p\u003e\n\u003cp\u003eTeam 2\u0026rsquo;s cohesion, therefore, was informed by the need to provide context for AI-generated output. That led to increased internal unity in coordinating the different tasks. As T2T3 explained, there was a consistent need to ensure that members charged with highlighting ST words and phrases with nuances provided the input necessary to reinforce the meaning of the TT, stating,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe needed to coordinate well with the members charged with underlining areas of the ST that carried the most meaning nuances. We knew that AI left out some meaning, maybe something that only humans could find. That was our job.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCommunication:\u003c/strong\u003e While communication remained important, its focus and nature shifted depending on the technology used in the translation. For Team 1, communication primarily centered on identifying errors, developing correction strategies, and ensuring consistency in addressing common MT flaws. As T1T5 noted,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMost of our discussions were about specific MT errors. We discussed the different error types, particularly those related to tenses, and found ways to resolve them. Often, we disagreed on how to resolve the problems, but we always ended up with a solution.\u003c/p\u003e\n\u003cp\u003eT1T1 remarked that \u0026ldquo;we set up a second shared document just for MT error patterns so we simultaneously see each other\u0026rsquo;s fixes and arrive at conclusions quickly.\u0026rdquo; The idea of setting up a second shared file specifically for debating and resolving MT errors was commended by all team members. P1T6 claimed, \u0026ldquo;By isolating the MT errors in a separate document, we could focus on the fixes. I think this helped us tremendously in meeting the deadline.\u0026rdquo;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUnlike Team 1, Team 2\u0026rsquo;s communication tended to be more interpretive, focusing on refining the AI\u0026rsquo;s output, discussing stylistic nuances, and aligning on subtle contextual adaptations. For example, T2T2 explained the team had a lengthy discussion on the \u0026ldquo;feel\u0026rdquo; of the translation. This was because \u0026ldquo;AI got most words, but we had to discuss how to make it sound natural for the target audience. We didn\u0026rsquo;t want the translation to read like AI wrote it.\u0026rdquo; Other members corroborated the assertion, with T2T5 maintaining, \u0026ldquo;in addition to ensuring the terminology was correct, we focused on making the conversation more \u0026lsquo;human.\u0026rsquo; I think we spent too much time on this aspect because overall, I think AI did a good job.\u0026rdquo; However, T2T3 felt that the discussion involved more brainstorming than PE, arguing that,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDiscussions involved more brainstorming than PE. We kept asking questions like \u0026ldquo;Does this convey the right meaning or tone?\u0026rdquo; rather than just \u0026ldquo;Is this grammatically correct?\u0026rdquo; I think the focus shifted a bit here, and we didn\u0026rsquo;t exercise our fluency or grammatical competency. Discussing both language and meaning may be more fulfilling, I think. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe study highlights specific post-translational challenges stemming from variations in MT and AI outputs. Team 1 encountered difficulties with Baidu\u0026rsquo;s inaccurate and literal translations, which necessitated extensive adjustments to grammar and terminology, disrupted their workflow, and led to frustration; however, it also promoted camaraderie among team members. Team 2 addressed issues related to Kimi AI\u0026rsquo;s fluent yet contextually flawed outputs, necessitating subtle adjustments for tone and consistency. Additionally, technical difficulties, such as slow computers, presented personal challenges. Team 1\u0026rsquo;s communication revolved around correcting errors and utilized a shared document for greater efficiency, while Team 2 prioritized interpretive discussions to improve contextual understanding and style. These interactions influenced team cohesion, as Team 1 bonded over common challenges, and Team 2 came together through joint problem-solving efforts.\u003c/p\u003e\n\u003cp\u003eThe findings are consistent with existing research on translation and team dynamics. Team 1 faced difficulties with error-prone MT outputs, necessitating substantial grammatical and terminological corrections. This resonates with the observations of O\u0026rsquo;Brien (2011) and Guerberof-Arenas (2013), who indicate that low-quality MT leads to increased PE effort and disrupts workflows. In contrast, Team 2 encountered issues with AI-generated outputs that are fluent but contextually inaccurate, aligning with the perspectives of Koponen (2016) and Bowker and Buitrago Ciro (2019). These challenges underscore the need for nuanced adjustments to enhance meaning and consistency. Moreover, Team 1\u0026rsquo;s camaraderie in stressful situations supports Hackman\u0026rsquo;s (2002) assertion that shared challenges foster unity, while Team 2\u0026rsquo;s collaborative problem-solving mirrors the insights of Risku and Dickinson (2017) regarding interpretive teamwork. The differing communication styles are notable, with Team 1 concentrating on error correction (Karamanis et al., 2011; Krings, 2001) and Team 2 focusing on stylistic nuances (LeBlanc, 2013; Olohan, 2021), highlighting the various demands of PE.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIntegrating quantitative and qualitative findings:\u0026nbsp;\u003c/strong\u003eEmploying a convergent parallel mixed-methods approach, quantitative (time, errors, and readability) and qualitative (stress and communication) data were integrated through a matrix, analyzed separately, and synthesized to triangulate insights. It was observed that Team 2 took longer for PE (60 minutes) compared to Team 1 (48 minutes), reflecting their qualitative emphasis on interpretive discussions (T2T3, T2T2), which was influenced by Kimi AI\u0026rsquo;s fluent yet contextually inaccurate output (T2T5). This was consistent with their lower error count (12 vs. 19), as stress from TDJs (T2T4) and a focus on meaning over grammar (T2T3) hindered fluent error detection, leading to a reduced performance score (92% vs. 95%). Team 1\u0026rsquo;s swift error correction (T1P6) and team spirit (T1T4) alleviated TDJ stress (T1T3), allowing for quicker completion and higher performance, despite the flaws in Baidu\u0026rsquo;s translation (T1T1).\u003c/p\u003e"},{"header":"5. Discussion","content":"\u003cp\u003e\u003cstrong\u003e5.1. Summary of Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study analyzed the PE of technical texts under TDJs, comparing Team 1 (Baidu MT, MT) with Team 2 (Kimi AI translation) using a convergent parallel mixed-methods approach. The quantitative results indicated that Team 2 required more time to post-edit Kimi AI’s output, which, while slightly more readable, demanded careful adjustments. Team 1 corrected a higher number of errors (19 vs. 12) due to Baidu’s output being more error-prone. However, Team 2 achieved a superior performance score (95% vs. 92%), reflecting Kimi AI’s advantages over Baidu in terms of fluency and accuracy, although both teams faced difficulties with terminology and institutional names. The increased resource consultation by Team 1 appears to correlate with the task’s difficulty, the specific challenges encountered, and the time invested. In contrast, Team 2 depended on the AI tool as a reference. Qualitative data revealed widespread TDJ stress, which affected their focus on quality, with both teams prioritizing speed. Team 1’s effective error correction and teamwork stood in contrast to Team 2’s interpretive dialogues and cooperative problem-solving. Both teams confronted challenges: Team 1 dealt with Baidu’s grammatical inconsistencies, while Team 1 grappled with Kimi’s “over-smoothness” and inconsistent terminology. Although both teams engaged in proactive planning, stress and technical difficulties disrupted their workflows.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.2. Implications for Translator Training\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe findings highlight important implications for translator training within the PATT framework, focusing on areas such as TDJ preparation, MTPE, AI integration, and strategies specific to different text types. First, Team 2’s extended 60-minute discussion of Kimi AI’s contextually flawed outputs highlights the necessity of training students in PE to handle AI-generated texts under time constraints. Training programs should replicate AI outputs to help develop skills in enhancing tone, style, and consistency, utilizing collaborative tools (e.g., cloud-based documents) for improved discussions and collaboration. This approach addresses the PATT gap related to AI integration, which aligns with industry trends where AI tools, such as ChatGPT and Kimi AI, are increasingly employed for drafting (Bowker \u0026amp; Buitrago Ciro, 2019). Activities could include timed PE exercises using AI translations of technical texts, allowing trainees to focus on identifying critical contextual errors rather than merely achieving fluency, thereby improving both efficiency and quality. Furthermore, training programs need to strike a balance between AI usage and traditional resources, such as online dictionaries, term banks, glossaries, websites, and corpora, which provide translators, particularly those in teams, with a wider array of options.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSecond, Team 1’s superior error correction (19 errors) and performance (92%) reveal that low-quality MT necessitates strong error detection skills. Training should incorporate simulations featuring error-prone MT outputs and promote collaboration through tools such as shared error logs (T1P6) during team-debriefing sessions (TDJs). This approach addresses the gaps in MTPE training, equipping students for real-life scenarios where MT errors, such as literal translations, frequently occur (Guerberof-Arenas, 2012; O’Brien, 2011). Engaging in role-based exercises, such as designating terminology specialists (T1T2), can boost team efficiency and ensure preparation for demanding professional situations.\u003c/p\u003e\n\u003cp\u003eThird, the pervasive TDJ stress affecting quality focus requires stress management training. Incorporating techniques such as mindfulness and time prioritization, along with stress measurement tools (e.g., NASA-TLX), can help students manage cognitive load (Sweller, 1988). This could address the PATT gap in TDJ preparation, enabling trainees to balance speed and quality under pressure, as stress diminishes focus on perfection. Simulations should mimic TDJ scenarios, using stress scales to quantify impacts, aligning with Lazarus and Folkman’s (1984) stress appraisal theory.\u003c/p\u003e\n\u003cp\u003eFourth, the shortcomings in terminology of both tools emphasize the necessity for training tailored to specific text types. Programs should incorporate terminology management with glossaries or databases, especially for technical documents, to bridge the PATT gap related to text-type emphasis. Practical exercises could include developing term bases for industry-specific language to improve the quality of professional translations (Koponen, 2016). Furthermore, the distinct dynamics within teams, such as Team 1’s rapport compared to Team 2’s collaborative approach, indicate that training should promote flexible roles in response to translation outcomes. By simulating a range of MT and AI tasks, trainees can learn to utilize collaborative skills, thus addressing the PATT gap concerning teamwork dynamics in TDJs (Hackman, 2002). By implementing these comprehensive strategies, PATT can enhance employability for graduates and effectively prepare trainees for the evolving needs of the translation field.\u003c/p\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eThis research examined the process of producing machine- and AI-translated technical documents under tight deadlines to assess their impact on efficiency and team dynamics. Utilizing a convergent parallel mixed-methods approach, 12 trainee translators were split into two groups within a simulated translation agency: Team 1 post-edited Baidu\u0026rsquo;s MT, while Team 2 worked on output from Kimi AI. The main findings reveal that Team 2 took longer (60 minutes vs. Team 1\u0026rsquo;s 48 minutes) due to interpretive discussions resulting from Kimi AI\u0026rsquo;s fluent but contextually flawed text, which was somewhat more readable. Both teams achieved high scores, with Team 2 edging slightly ahead (95% vs. 92%). Team 1 made more corrections, applying effective error correction strategies and team spirit to tackle Baidu\u0026rsquo;s numerous mistakes, characterized by grammatical errors. Both tools faced difficulties with terminology and institutional names; Team 1 sought additional resources to overcome Baidu\u0026rsquo;s issues, whereas Team 2 depended significantly on the AI tool. The considerable stress from TDJs, as reported by both teams, led to a focus on speed over quality, with technical issues further hindering their workflows.\u003c/p\u003e\u003cp\u003eThese results have important implications for translator education within the PATT framework. Training programs need to mimic AI and MT PE in TDJs, guiding trainees in refining AI outputs for tone and consistency while identifying errors in subpar MT through collaborative tools such as shared error logs. Techniques for managing stress, aided by tools such as stress scales, can alleviate pressure related to TDJs, improving focus on quality. Effective terminology management, utilizing glossaries, is crucial for technical documents. Promoting adaptive team roles also enhances collaboration and communication. These strategies tackle PATT shortcomings in TDJ preparation, MTPE, AI integration, and text-specific training, equipping graduates for the demanding and evolving translation industry.\u003c/p\u003e"},{"header":"7. Limitations","content":"\u003cp\u003eThe small sample size of 12 trainee translators in this study limits the ability to generalize findings to larger populations or professional contexts. The use of a single technical news article restricts insights into various text types, such as legal or literary translations, which may lead to missing genre-specific PE challenges. While the 80-minute simulation aims to reflect tight deadlines, it may not accurately represent the ongoing pressure of actual translation workflows, potentially underestimating stress effects. The reliance on self-reported reflections for qualitative data may introduce bias, as trainees could either exaggerate or downplay their experiences. Uncontrolled variables, such as differences in team composition, prior knowledge of Baidu or Kimi AI tools, or individual language skills, could affect results and comparisons between teams. Additionally, focusing on one MT and one AI tool limits the study\u0026rsquo;s conclusions regarding other tools. Future studies with larger samples, varied texts, and objective measurements could help address these limitations, thus increasing the applicability of PATT training.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eDisclosure statement\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eThe authors report there are no competing interests to declare\u003c/em\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData generated during this research is available with the corresponding author upon reasonable request.\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Guangdong Province Postgraduate Education Innovation Project under Grant 2023JGXM-066.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Approval\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was conducted under the guidelines of the Helsinki Declaration and with approval from the ethics department of the School of Foreign Languages, Guangdong University of Finance and Economics (Approval No. SFLGDUFE202501211 of Jan. 21, 2025).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInformed consent\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWritten informed consent was obtained from participants on January 26, 2025, prior to the study.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eApandi, A., \u0026amp; Afiah, D. S. S. (2019). Project based learning in translation class. \u003cem\u003eAcademic Journal Perspective: Education, Language, and Literature\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(2), 101-108. https://doi.org/10.33603/perspective.v7i2.2656 \u003c/li\u003e\n\u003cli\u003eApandi, A., \u0026amp; Afiah, D. S. S. (2019). Project based learning in translation class. \u003cem\u003eAcademic Journal Perspective: Education, Language, and Literature\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(2), 101-108.\u003c/li\u003e\n\u003cli\u003eApandi, Apandi, and Devi Siti Sihatul Afiah. \u0026quot;Project based learning in translation class.\u0026quot; \u003cem\u003eAcademic Journal Perspective: Education, Language, and Literature\u003c/em\u003e 7.2 (2019): 101-108. \u003c/li\u003e\n\u003cli\u003eBowker, L. (2004). What does it take to work in the translation profession in Canada in the 21st century? Exploring a database of job advertisements. \u003cem\u003eMeta, 49\u003c/em\u003e(1), 960\u0026ndash;972. \u003c/li\u003e\n\u003cli\u003eBowker, L., \u0026amp; Buitrago Ciro, J. (2019). \u003cem\u003eMachine translation and global research: Towards improved machine translation literacy in the scholarly community\u003c/em\u003e. Emerald Publishing.\u003c/li\u003e\n\u003cli\u003eBundgaard, K. (2017). Translator attitudes towards translator-computer interaction\u0026ndash; Findings from a workplace study. \u003cem\u003eHermes\u003c/em\u003e, (56), 125-144. https://doi.org/10.7146/hjlcb.v0i56.97228 \u003c/li\u003e\n\u003cli\u003eBuysschaert, J., Fern\u0026aacute;ndez-Parra, M., Kerremans, K., Koponen, M., \u0026amp; Van Egdom, G. W. (2018). Embracing digital disruption in translator training: Technology immersion in simulated translation bureaus. \u003cem\u003eTradum\u0026agrave;tica\u003c/em\u003e, (16), 0125-133. https://doi.org/10.5565/rev/tradumatica.209 \u003c/li\u003e\n\u003cli\u003eCarl, M., Gutermuth, S., \u0026amp; Hansen-Schirra, S. (2015). Post-editing machine translation: A usability test for professional translation settings. In \u003cem\u003ePsycholinguistic and cognitive inquiries into translation and interpreting\u003c/em\u003e (pp. 145-174). John Benjamins Publishing Company.\u003c/li\u003e\n\u003cli\u003eChesterman, A. (2005). Problems with strategies. In K. K\u0026aacute;roly \u0026amp; \u0026Aacute;. F\u0026oacute;ris (Eds.), \u003cem\u003eNew trends in translation studies: In honour of Kinga Klaudy\u003c/em\u003e (pp. 17\u0026ndash;28). Akad\u0026eacute;miai Kiad\u0026oacute;.\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a Gonz\u0026aacute;lez, M., \u0026amp; Veiga D\u0026iacute;az, M. T. (2015). Guided Inquiry and Project-Based Learning in the field of specialised translation: a description of two learning experiences. \u003cem\u003ePerspectives\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(1), 107-123. \u003c/li\u003e\n\u003cli\u003eGonz\u0026aacute;lez-Davies, M., \u0026amp; Ra\u0026iacute;do, V. E. (2018). Situated learning in translator and interpreter training: Bridging research and good practice. In \u003cem\u003eSituated Learning in translator and interpreter training\u003c/em\u003e (pp. 1-11). Routledge. \u003c/li\u003e\n\u003cli\u003eGouadec, D. (2007). \u003cem\u003eTranslation as a profession\u003c/em\u003e. Amsterdam, Netherlands: John Benjamins. \u003c/li\u003e\n\u003cli\u003eGuerberof-Arenas, A. (2013). The role of post-editing in translation workflows: Productivity and quality. \u003cem\u003eThe Journal of Specialised Translation, 20\u003c/em\u003e, 76\u0026ndash;92.\u003c/li\u003e\n\u003cli\u003eHackman, J. R. (2002). \u003cem\u003eLeading teams: Setting the stage for great performances\u003c/em\u003e. Harvard Business Review Press. \u003c/li\u003e\n\u003cli\u003eHackman, J. R. (2002). \u003cem\u003eLeading teams: Setting the stage for great performances\u003c/em\u003e. Harvard Business Review Press.\u003c/li\u003e\n\u003cli\u003eHariyanto, S. (2016). Assessment in Translation Research, Teaching, and Industry. In \u003cem\u003eThe 2016 International Translation and Interpreting Symposium\u003c/em\u003e (pp. 236-244). Karamanis, N., Luz, S., \u0026amp; Doherty, G. (2011). Translation practice in the age of computer-assisted translation: The impact of technology on workflow. \u003cem\u003eThe Translator, 17\u003c/em\u003e(2), 231\u0026ndash;250. https://doi.org/10.1080/13556509.2011.10799487\u003c/li\u003e\n\u003cli\u003eKiraly, D. (2000). \u003cem\u003eA social constructivist approach to translator education: Empowerment from theory to practice\u003c/em\u003e. Manchester, England: St. Jerome.\u003c/li\u003e\n\u003cli\u003eKiraly, D. (2005). Project-based learning: A case for situated translation. \u003cem\u003eMeta\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(4), 1098-1111. https://doi.org/10.7202/012063ar \u003c/li\u003e\n\u003cli\u003eKiraly, Don. \u0026quot;Project-based learning: A case for situated translation.\u0026quot; \u003cem\u003eMeta\u003c/em\u003e 50.4 (2005): 1098-1111. \u003c/li\u003e\n\u003cli\u003eKonttinen, K., Veivo, O., \u0026amp; Salo, P. (2019). Translation students\u0026rsquo; conceptions of translation workflow in a simulated translation company environment. \u003cem\u003eThe Interpreter and Translator Trainer\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 79\u0026ndash;94. https://doi.org/10.1080/1750399X.2019.1619218 \u003c/li\u003e\n\u003cli\u003eKoponen, M. (2010). Assessing machine translation quality with error analysis. \u003cem\u003eMikael: K\u0026auml;\u0026auml;nt\u0026auml;misen ja tulkkauksen tutkimuksen aikakauslehti\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e. Koponen, M. (2016). Is machine translation post-editing worth the effort? A survey of research into post-editing and effort. \u003cem\u003eThe Journal of Specialised Translation, 25\u003c/em\u003e, 131\u0026ndash;148.\u003c/li\u003e\n\u003cli\u003eKozyar, M. M., Pasichnyk, S. M., Kopchak, M. M., Burmakina, N. S., \u0026amp; Suran, T. (2022). Simulation-Based Learning as an Effective Method of Practical Training of Future Translators. \u003cem\u003eJournal of Curriculum and teaching\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 298-308. Krings, H. P. (2001). \u003cem\u003eRepairing texts: Empirical investigations of machine translation post-editing processes\u003c/em\u003e. Kent State University Press.\u003c/li\u003e\n\u003cli\u003eK\u0026uuml;nzli, A. (2004). Risk taking: Trainee translators vs. professional translators. A case study. \u003cem\u003eThe Journal of Specialised Translation, 2\u003c/em\u003e, 34\u0026ndash;49.\u003c/li\u003e\n\u003cli\u003eKussmaul, P. (1995). \u003cem\u003eTraining the translator\u003c/em\u003e. Amsterdam, Netherlands: John Benjamins.\u003c/li\u003e\n\u003cli\u003eLazarus, R. S., \u0026amp; Folkman, S. (1984). \u003cem\u003eStress, appraisal, and coping\u003c/em\u003e. Springer Publishing Company. \u003c/li\u003e\n\u003cli\u003eLeBlanc, M. (2013). Translators\u0026rsquo; perceptions of machine translation: A Canadian survey. \u003cem\u003eThe Interpreter and Translator Trainer, 7\u003c/em\u003e(2), 193\u0026ndash;214. https://doi.org/10.1080/1750399X.2013.845209\u003c/li\u003e\n\u003cli\u003eLi, D., Zhang, C., \u0026amp; He, Y. (2015). Project-based learning in teaching translation: students\u0026rsquo; perceptions. \u003cem\u003eThe Interpreter and Translator Trainer\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(1), 1-19. https://doi.org/10.1080/1750399X.2015.1010357\u003c/li\u003e\n\u003cli\u003eMarks, M. A., Mathieu, J. E., \u0026amp; Zaccaro, S. J. (2001). A temporally based framework and taxonomy of team processes. \u003cem\u003eAcademy of Management Review, 26\u003c/em\u003e(3), 356\u0026ndash; 376. https://doi.org/10.5465/amr.2001.4845785\u003c/li\u003e\n\u003cli\u003eMassey, G., \u0026amp; Ehrensberger-Dow, M. (2017). Machine translation and cognitive load: Insights from eye-tracking and retrospective protocols. \u003cem\u003eThe Journal of Specialised Translation, 28\u003c/em\u003e, 138\u0026ndash;155. \u003c/li\u003e\n\u003cli\u003eMcDermid, C. (2025). Allyship and social annotation software in interpreter education. \u003cem\u003eThe Interpreter and Translator Trainer\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(2), 137-154. https://doi.org/10.1080/1750399X.2025.2488193 \u003c/li\u003e\n\u003cli\u003eMitchell-Schuitevoerder, R. (2013, October). A project-based methodology in translator training. In \u003cem\u003eTracks and Treks in Translation Studies: Selected Papers from the EST Congress, Leuven 2010\u003c/em\u003e (pp. 127-142). John Benjamins Publishing Company. \u003c/li\u003e\n\u003cli\u003eMo, A., \u0026amp; Man, D. (2017). The ecosystem of translator workstation: Learning electronic tools in a training program for professional translators in China. \u003cem\u003eBabel\u003c/em\u003e, \u003cem\u003e63\u003c/em\u003e(3), 401-422. https://doi.org/10.1075/babel.63.3.06aip \u003c/li\u003e\n\u003cli\u003eMor\u0026oacute;n, M., \u0026amp; Calvo, E. (2018). Introducing transcreation skills in translator training contexts: A situated project-based approach. \u003cem\u003eThe Journal of Specialised Translation\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e, 126-148. \u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Brien, S. (2011). Towards predicting post-editing productivity. \u003cem\u003eMachine Translation, 25\u003c/em\u003e(3), 197\u0026ndash;215. https://doi.org/10.1007/s10590-011-9096-2\u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Brien, S. (2021). Post-editing. In \u003cem\u003eHandbook of translation studies\u003c/em\u003e (pp. 177-183). John Benjamins Publishing Company. \u003c/li\u003e\n\u003cli\u003eOlohan, M. (2021). \u003cem\u003eTranslation and technology: The evolution of translation practice\u003c/em\u003e. Routledge. \u003c/li\u003e\n\u003cli\u003eOlvera Lobo, M. D., Castro Prieto, M. R., Quero Gervilla, E., Mu\u0026ntilde;oz Mart\u0026iacute;n, R., Mu\u0026ntilde;oz Raya, E., Murillo Melero, M., ... \u0026amp; Dom\u0026iacute;nguez L\u0026oacute;pez, C. (2005). Translator training and modern market demands. \u003cem\u003ePerspectives\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(2), 132-142. https://cris.unibo.it/handle/11585/697109 \u003c/li\u003e\n\u003cli\u003eOlvera Lobo, M. D., Robinson, B., Castro Prieto, R. M., Quero Gervilla, E., Mu\u0026ntilde;oz Mart\u0026iacute;n, R., Mu\u0026ntilde;oz Raya, E., ... \u0026amp; D\u0026iacute;ez Lerma, J. L. (2007). A professional approach to translator training (PATT). \u003cem\u003eMeta\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(3), 517-528. https://doi.org/10.7202/016736ar \u003c/li\u003e\n\u003cli\u003ePark, D., \u0026amp; Pad\u0026oacute;, S. (2024). Multi-Dimensional Machine Translation Evaluation: Model Evaluation and Resource for Korean https://doi.org/10.48550/arXiv.2403.12666 \u003c/li\u003e\n\u003cli\u003ePrieto-Velasco, J. A., \u0026amp; Fuentes-Luque, A. (2018). A collaborative multimodal working environment for the development of instrumental and professional competences of student translators: an innovative teaching experience. In \u003cem\u003eSituated Learning in Translator and Interpreter Training\u003c/em\u003e (pp. 76-91). Routledge. \u003c/li\u003e\n\u003cli\u003ePym, A. (2009). Translator training. In \u003cem\u003eOxford companion to translation studies\u003c/em\u003e (Pre- print). Tarragona, Spain: Intercultural Studies Group. \u003c/li\u003e\n\u003cli\u003ePym, A. (2009). Translator training. \u003cem\u003eProyecto para El Oxford companion translation studies\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eRisku, H., \u0026amp; Dickinson, A. (2017). Translation as teamwork: A situated approach to translators\u0026rsquo; collaboration. \u003cem\u003eTranslation Spaces, 6\u003c/em\u003e(1), 81\u0026ndash;99. https://doi.org/10.1075/ts.6.1.05ris\u003c/li\u003e\n\u003cli\u003eRisku, H., \u0026amp; Dickinson, A. (2017). Translation as teamwork: A situated approach to translators\u0026rsquo; collaboration. \u003cem\u003eTranslation Spaces, 6\u003c/em\u003e(1), 81\u0026ndash;99. https://doi.org/10.1075/ts.6.1.05ris\u003cu\u003e\u003cbr\u003e\u003c/u\u003eSalas, E., Shuffler, M. L., Thayer, A. L., \u0026amp; Bedwell, W. L. (2015). Understanding and improving teamwork in organizations: A scientifically based practical guide. \u003cem\u003eHuman Resource Management, 54\u003c/em\u003e(4), 599\u0026ndash;622. https://doi.org/10.1002/hrm.21628\u003c/li\u003e\n\u003cli\u003eSweller, J. (1988). Cognitive load during problem solving: Effects on learning. \u003cem\u003eCognitive Science, 12\u003c/em\u003e(2), 257\u0026ndash;285. https://doi.org/10.1207/s15516709cog1202_4 Tekwa, K., Su, W., \u0026amp; Li, D. (2024). Web 2.0 technologies and translator training: assessing trainees\u0026rsquo; use of instant messaging as a collaborative tool in accomplishing translation tasks. \u003cem\u003eHumanities and Social Sciences Communications\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 1-14. https://doi.org/10.1057/s41599-024-02934-5 \u003c/li\u003e\n\u003cli\u003eTsai, Y. (2025). Tuning language and lyrics: a case study of song translation in the translator training classroom. \u003cem\u003eThe Interpreter and Translator Trainer\u003c/em\u003e, 1-19. https://doi.org/10.1080/1750399X.2025.2507541\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Machine translation post-editing, AI translation, Tight deadline job, Professional approach to translator training, Team dynamics, Translator training","lastPublishedDoi":"10.21203/rs.3.rs-6892673/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6892673/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis research investigates the post-editing (PE) of machine- and artificial intelligence (AI)-generated technical documents in the context of tight deadline jobs (TDJs) to assess team dynamics and efficiency. In a mock translation agency setting, 12 trainee translators were divided into two groups: Team 1 focused on PE Baidu\u0026rsquo;s machine translation (MT), while Team 2 worked on the outputs from Kimi AI, following a convergent parallel mixed-methods design. Quantitative metrics monitored time, errors, and consulted resources, while qualitative insights explored stress levels and teamwork. Results indicate that Team 1 invested more time (60 vs. 48 minutes) and corrected more errors (19 vs. 12) due to the prevalence of mistakes in Baidu\u0026rsquo;s output. They also consulted more resources but achieved lower accuracy (92% vs. 95%) than Team 2, which improved Kimi AI\u0026rsquo;s fluent yet contextually flawed translations. Both teams tended to prioritize speed over quality, with ongoing challenges in terminology. The study's implications for the professional approach to translator training (PATT) involve simulating AI/MT PE to improve error detection, making contextual adjustments, incorporating stress management strategies, and promoting terminology proficiency and flexible team roles. This research helps address gaps in PATT related to TDJ training, machine-translated post-editing, AI functionality integration, and specialized text training, thereby enhancing translator education to align with current industry needs.\u003c/p\u003e","manuscriptTitle":"Team dynamics and efficiency in post-editing: Machine translation systems versus artificial intelligence chatbots under tight deadline conditions","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-25 10:02:48","doi":"10.21203/rs.3.rs-6892673/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9bcd8c06-4495-45e3-8c62-bbc14aa791e5","owner":[],"postedDate":"September 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55167835,"name":"Humanities/Language and linguistics"},{"id":55167836,"name":"Social science/Education"},{"id":55167837,"name":"Social science/Language and linguistics"},{"id":55167838,"name":"Social science/Social policy"},{"id":55167839,"name":"Social science/Sociology"}],"tags":[],"updatedAt":"2025-12-15T11:25:02+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-25 10:02:48","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6892673","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6892673","identity":"rs-6892673","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}