Earticle

Generative AI has revolutionized machine translation (MT) by enabling customizable outputs through prompt engineering. Unlike traditional neural machine translation (NMT) systems such as Google Translate and DeepL, which rely on predefined translation patterns, models like GPT-4 allow users to refine translation quality through different prompting strategies. This study examines the impact of Zero-shot, Few-shot, Style-specific, and Step-by-step prompting on translation quality and evaluates the effectiveness of Auto Prompt Optimization (APO) in further enhancing performance. Using GPT-4, translations were conducted for Korean-to-English, Japanese, Chinese (Ko→En, Ja, Zh) language pairs. The study assessed BLEU, TER, and COMET scores to measure accuracy, fluency, and semantic consistency. Results show that Few-shot and Step-by-step prompting significantly improve translation quality, while APO further enhances BLEU scores and reduces TER values across all prompting types. Findings also indicate that prompt effectiveness varies by language pair, emphasizing the need for language-specific prompt strategies. This study provides empirical evidence of how prompt engineering influences AI translation quality and offers strategic guidelines for optimizing translation performance.

This study discusses how media consumers, particularly movie fans, use online platforms to exchange and contest ideas, debate, and enhance their understanding of language and culture in response to information loss in multilingual film subtitles. While subtitling inevitably involves some degree of information loss, this issue is especially significant in multilingual films, where language plays a vital role in character development and narrative progression. Using Quentin Tarantino’s Inglourious Basterds (2009) as a case study and drawing on Henry Jenkins’ concept of participatory culture (2006) and Pierre Lévy’s idea of collective intelligence, this study examines how media audiences engage with language, culture, and subtitles within and across different fan communities and digital platforms. The findings reveal that media consumers actively share, test, reaffirm, and acquire missing linguistic and cultural knowledge related to multilingual scenes through a decentralized, collaborative, and democratic process.

This study focuses on evaluating Chinese interpretations of English questions posed by international journalists during China's Premier Press Conferences (PPCs) from 2018 to 2022, to examine the function of interpreting as a gatekeeping mechanism in mediating political discourse. By utilizing the Critical Discourse Analysis (CDA) approach, the discrepancies identified in target texts are analyzed based on the adopted translation strategies: addition, omission, and modification. The findings reveal that interpreting is a significant tool for managing linguistic and ideological nuances, and it plays an important role in shaping domestic and international perceptions of China. In the context of political communication, strategies adopted by interpreters are influenced by the government's perspectives, aim to mitigate adversarial expressions, and reinforce a favorable national image. Additionally, the analysis of the interplay between language, power, and ideology provides evidence for the necessity of further interpreting and translation research in diplomatic contexts.

This study explores the social aspect of translation from the perspective of a sociology of translation. A sociology of translation shifts the focus from texts to the sociocultural contexts, the reasons conditioning a translation process, social networks and agents. This study traces the social aspect of retranslation of a ‘screenseller’ in the production of the novel Pachinko by analyzing transformation of the field of publishing. In Korea, the retranslated version of Pachinko was produced in 2022, based on the pursuit of economic capital of a publishing company. While Pachinko enjoyed a status of ‘screenseller’ which relied on video production, many publishing companies competed with each other to own the copyright for the book which would lead it to retranslation. To investigate the social aspect of retranslation implied in the production of the cultural product, this study applies Pierre Bourdieu’s field theory of cultural production and analyzes newspaper articles. The retranslation of Pachinko was conducted by the publishing company, Influential which won the competition in the indirect contexts of the emergence of ‘New generation’, born in 1970s and the cultural phenomenon of ‘transmedia storytelling’. Influential produced a retranslated version of Pachinko with detection of public reader’s dispositions, while using proper marketing strategy as retranslation. This study reasons out the economic and sociocultural practice of retranslation from the production of the retranslated version of the ‘screenseller’.

This study analyzes the translation activities of TeamWaldo, a user-driven game translation community in Korea, as a case study to explore how user-participation translation takes place in the digital age. Focusing on TeamWaldo’s collaborative translation practices, the study examines Team Waldo translators’ motivations, decision-making processes, collaboration methods, as well as the reception of their translations among game players. The findings indicate that Team Waldo engages in community-based and cooperative translation, characterized by active discussions, feedback exchanges, and ongoing revisions. Translators prioritize user experience and immersion in the game over language conversion, often adapting humor, wordplay, and cultural elements. They also adopt a user-friendly approach through their “brand,” built on the past successful experiences of unofficial Korean versions and the team’s prestige in the gaming community. This study also highlights translation as a collective and interactive process in the digital age. Translation decisions, such as term choices, are determined through community voting, demonstrating a participatory approach. By shedding light on the role and significance of user-participation translation in game translation, this study contributes to understanding the evolving nature of translation in the digital era and its potential for broader applications.

This study employs Actor-Network Theory (ANT) to analyze two English translations of The Vegetarian, examining how various human and non-human actors interact to establish translation networks. Janet Hong’s translation, intended to introduce Korean literature to the English-speaking world, faced challenges due to the absence of publisher enrollment within the network. In contrast, Deborah Smith’s translation successfully established a robust actor-network by engaging the publisher, agents, and the author, while strategically leveraging non-human actors—such as the London Book Fair and sample translations—to secure a publishing contract. The findings indicate that the success of a translation is not solely dependent on the translator’s individual expertise but also on the effective collaboration and stability of a diverse network of actors. This study further underscores the translator's role as an active actor who not only conveys textual meaning but also reshapes and enhances value within networks. For Korean literature to achieve international recognition, it is essential for actors to actively construct networks within the target culture and strategically mobilize non-human actors. By employing a sociological approach to translation, specifically through the application of Actor-Network Theory, this study contributes to translation studies by illuminating the complex social dynamics involved in literary translation processes, thus providing significant insights into the global circulation of Korean literature.

This study explores the feasibility of incorporating cohesion as an evaluation criterion beyond the sentence level inmachine translation assessment. To this end, the study employed Coh-Metrix to measure the cohesion of English translations produced by Google Translate, DeepL, and ChatGPT for 68 Korean editorials, focusing on three criteria: pronouns, connectives, and repetition. One-way ANOVA and Tukey’s HSD test were conducted to determine whether there were significant differences in cohesion among the machine translation engines. The results revealed differences in content word overlap, latent semantic cohesion between adjacent sentences, and the distribution of given and new information across translation outputs. Additionally, variations were observed in the frequency of temporal and additive connectives, as well as in the occurrence of first-person singular and plural pronouns. A detailed analysis of cases with significant differences showed that, while individual sentences may appear accurate, the translations sometimes failed to convey the original meaning accurately within a larger context, potentially hindering reader comprehension. This study is significant in that it moves beyond traditional sentence-level evaluations and suggests the potential for incorporating cohesion as a criterion in machine translation assessment.