Hasil untuk "Translating and interpreting"

Gabriel Maldonado Pantoja

Este trabajo, como iniciativa docente independiente, propone integrar la ética profesional y el activismo social en la formación de traductores médicos. Desde los principios de Freire y las contribuciones de Giroux, así como de Tymoczko, Valero-Garcés y Tipton, se plantea la traducción médica como una práctica social que promueve justicia y equidad en contextos vulnerables. Así, el objetivo fue analizar la evolución en la percepción de la ética en la traducción médica por parte de los estudiantes, mediante el diseño, la implementación y la evaluación de estrategias didácticas basada en la pedagogía crítica y el activismo, orientadas a fortalecer su formación ética y activista. La investigación se desarrolló en dos fases. En la primera se aplicó una encuesta cualitativa a estudiantado que cursó Traducción Médica bajo un plan de estudios anterior, donde la carta descriptiva aludía a la responsabilidad social del traductor, pero sin abordarla en los contenidos o actividades. En la segunda, con un nuevo grupo y un plan actualizado que incluye la ética como eje desde el primer módulo, se diseñó e implementó una intervención de cuatro semanas, con actividades sobre perfil profesional, dilemas bioéticos y creación de materiales accesibles. Los resultados de la fase 1 mostraron que el estudiantado reconoce la ética y la precisión terminológica como aspectos importantes, aunque de forma general. En la fase 2 se observó una evolución más profunda, donde reconocieron la precisión terminológica como eje ético, asumieron mayor conciencia sobre su impacto en la accesibilidad lingüística, comprendieron las implicaciones legales de su labor y expresaron mayor seguridad para enfrentar dilemas éticos. Esta propuesta refuerza el papel del traductor médico como agente de cambio social.

Ido Levy, Orr Paradise, Boaz Carmeli et al.

Emergent Communication (EC) provides a unique window into the language systems that emerge autonomously when agents are trained to jointly achieve shared goals. However, it is difficult to interpret EC and evaluate its relationship with natural languages (NL). This study employs unsupervised neural machine translation (UNMT) techniques to decipher ECs formed during referential games with varying task complexities, influenced by the semantic diversity of the environment. Our findings demonstrate UNMT's potential to translate EC, illustrating that task complexity characterized by semantic diversity enhances EC translatability, while higher task complexity with constrained semantic variability exhibits pragmatic EC, which, although challenging to interpret, remains suitable for translation. This research marks the first attempt, to our knowledge, to translate EC without the aid of parallel data.

Zhong Zhou

Translation into severely low-resource languages has both the cultural goal of saving and reviving those languages and the humanitarian goal of assisting the everyday needs of local communities that are accelerated by the recent COVID-19 pandemic. In many humanitarian efforts, translation into severely low-resource languages often does not require a universal translation engine, but a dedicated text-specific translation engine. For example, healthcare records, hygienic procedures, government communication, emergency procedures and religious texts are all limited texts. While generic translation engines for all languages do not exist, translation of multilingually known limited texts into new, low-resource languages may be possible and reduce human translation effort. We attempt to leverage translation resources from rich-resource languages to efficiently produce best possible translation quality for well known texts, which are available in multiple languages, in a new, low-resource language. To reach this goal, we argue that in translating a closed text into low-resource languages, generalization to out-of-domain texts is not necessary, but generalization to new languages is. Performance gain comes from massive source parallelism by careful choice of close-by language families, style-consistent corpus-level paraphrases within the same language and strategic adaptation of existing large pretrained multilingual models to the domain first and then to the language. Such performance gain makes it possible for machine translation systems to collaborate with human translators to expedite the translation process into new, low-resource languages.

Yunmeng Li, Jun Suzuki, Makoto Morishita et al.

Machine translation models are still inappropriate for translating chats, despite the popularity of translation software and plug-in applications. The complexity of dialogues poses significant challenges and can hinder crosslingual communication. Instead of pursuing a flawless translation system, a more practical approach would be to issue warning messages about potential mistranslations to reduce confusion. However, it is still unclear how individuals perceive these warning messages and whether they benefit the crowd. This paper tackles to investigate this question and demonstrates the warning messages' contribution to making chat translation systems effective.

Sara Rezaeimanesh, Faezeh Hosseini, Yadollah Yaghoobzadeh

Large language models (LLMs) have shown superior capabilities in translating figurative language compared to neural machine translation (NMT) systems. However, the impact of different prompting methods and LLM-NMT combinations on idiom translation has yet to be thoroughly investigated. This paper introduces two parallel datasets of sentences containing idiomatic expressions for Persian$\rightarrow$English and English$\rightarrow$Persian translations, with Persian idioms sampled from our PersianIdioms resource, a collection of 2,200 idioms and their meanings, with 700 including usage examples. Using these datasets, we evaluate various open- and closed-source LLMs, NMT models, and their combinations. Translation quality is assessed through idiom translation accuracy and fluency. We also find that automatic evaluation methods like LLM-as-a-judge, BLEU, and BERTScore are effective for comparing different aspects of model performance. Our experiments reveal that Claude-3.5-Sonnet delivers outstanding results in both translation directions. For English$\rightarrow$Persian, combining weaker LLMs with Google Translate improves results, while Persian$\rightarrow$English translations benefit from single prompts for simpler models and complex prompts for advanced ones.

Ergun Biçici

We use transductive regression techniques to learn mappings between source and target features of given parallel corpora and use these mappings to generate machine translation outputs. We show the effectiveness of $L_1$ regularized regression (\textit{lasso}) to learn the mappings between sparsely observed feature sets versus $L_2$ regularized regression. Proper selection of training instances plays an important role to learn correct feature mappings within limited computational resources and at expected accuracy levels. We introduce \textit{dice} instance selection method for proper selection of training instances, which plays an important role to learn correct feature mappings for improving the source and target coverage of the training set. We show that $L_1$ regularized regression performs better than $L_2$ regularized regression both in regression measurements and in the translation experiments using graph decoding. We present encouraging results when translating from German to English and Spanish to English. We also demonstrate results when the phrase table of a phrase-based decoder is replaced with the mappings we find with the regression model.

Erik Coenen

Este estudio pretende examinar y comparar las estrategias traductoras de las cinco traducciones conocidas de La vida es sueño al portugués, sobre todo en lo que se refiere al tratamiento que se le brinda a la forma poética. Por ello, presta más atención a las traducciones en verso que a las dos que adoptan la prosa. Por medio del análisis individual y comparativo del monólogo inicial de la obra en las cinco traducciones, aspira asimismo a reconstruir hasta cierto punto el proceso de traducción en verso, y así arrojar luz sobre la traducción en verso rimado en general.

Laura Theys, Cornelia Wermuth, Heidi Salaets et al.

Current interpreter training programs pay increasingly more attention to the intricacies of the clinical context, such as doctors and patients’ communicative goals. However, to date, the conduit model remains influential when it comes to interpreters dealing with other participants’ emotions and their own emotions in interpreter-mediated consultations (IMCs). Consequently, establishing a good doctor-patient relationship by means of empathic communication (EC) might be jeopardized in IMCs. During EC, patients express their emotional or illness experiences to which doctors convey their empathic understanding. This study aimed to assess how doctors, patients, and interpreters verbally co-construct EC and the interpreter’s effect on this process. We analyzed 7 authentic IMCs using the Empathic Communication Coding System, as previously adapted for IMCs. We identified empathic opportunities (EOs) and empathic responses (ERs) as expressed by patients/doctors, and as rendered by interpreters. Our results showed that EC is the result of an interactive and collaborative process among all participants in IMCs. That is, the interplay between participants’ communicative actions determines how patients’ expressed lived experiences are addressed in IMCs. Our findings suggest that interpreters hold a central position in this process as they initiated EC about the patient’s illness experience and exerted control over the ways in which statements were rendered (e.g., interpreters omitted and altered original statements). In addition, our results indicated that EC in IMCs might be compromised by doctors and interpreters’ communicative actions.

Prottay Kumar Adhikary, Bandaru Sugandhi, Subhojit Ghimire et al.

In today's globalized world, effective communication with people from diverse linguistic backgrounds has become increasingly crucial. While traditional methods of language translation, such as written text or voice-only translations, can accomplish the task, they often fail to capture the complete context and nuanced information conveyed through nonverbal cues like facial expressions and lip movements. In this paper, we present an end-to-end video translation system that not only translates spoken language but also synchronizes the translated speech with the lip movements of the speaker. Our system focuses on translating educational lectures in various Indian languages, and it is designed to be effective even in low-resource system settings. By incorporating lip movements that align with the target language and matching them with the speaker's voice using voice cloning techniques, our application offers an enhanced experience for students and users. This additional feature creates a more immersive and realistic learning environment, ultimately making the learning process more effective and engaging.

J. Reimand, Tambet Arak, P. Adler et al.

Functional enrichment analysis is a key step in interpreting gene lists discovered in diverse high-throughput experiments. g:Profiler studies flat and ranked gene lists and finds statistically significant Gene Ontology terms, pathways and other gene function related terms. Translation of hundreds of gene identifiers is another core feature of g:Profiler. Since its first publication in 2007, our web server has become a popular tool of choice among basic and translational researchers. Timeliness is a major advantage of g:Profiler as genome and pathway information is synchronized with the Ensembl database in quarterly updates. g:Profiler supports 213 species including mammals and other vertebrates, plants, insects and fungi. The 2016 update of g:Profiler introduces several novel features. We have added further functional datasets to interpret gene lists, including transcription factor binding site predictions, Mendelian disease annotations, information about protein expression and complexes and gene mappings of human genetic polymorphisms. Besides the interactive web interface, g:Profiler can be accessed in computational pipelines using our R package, Python interface and BioJS component. g:Profiler is freely available at http://biit.cs.ut.ee/gprofiler/.

Alba Fernández Santana, Celia Martín de León

Desde una perspectiva corpórea de la cognición, los gestos representacionales se han descrito como creaciones espontáneas que emergen de la producción de imágenes mentales durante los procesos de construcción de significados. El objetivo de este trabajo es explorar el papel que desempeñan este tipo de gestos en los procesos de construcción de significados de los intérpretes simultáneos. Con este fin, se han estudiado las relaciones entre los gestos icónicos realizados espontáneamente por cuatro intérpretes en cabina y las imágenes mentales que recordaban haber producido durante la interpretación. Los resultados ofrecen indicios convergentes de una vinculación entre los gestos analizados y las imágenes mentales descritas por las participantes y permiten formular algunas hipótesis sobre el origen y las funciones de los gestos icónicos producidos por los intérpretes durante la interpretación simultánea.

Forough Farahmand Haromi, Hossein Mohtadi, Mohammad Javad Pourabed

The style of exaggeration, as one of the most frequent syntactic styles, with many latent meanings, has always been the focus of translators. Examining the translation of this style in Nahj al-Balagha, which is a book mixed with eloquent expressions, is very important. On the other hand, Catford, as a theoretician in translation who has presented a precise and meticulous theory in translation, can be a good criterion and scale for analyzing the meaning of translations of Nahj al-Balagha. Be exaggerated in style. This style is used in morphological, syntactic, and rhetorical structures; but since one of the most used of this style is the modified weights of the subject noun, this research has tried to extract these weights with a descriptive-analytical method and apply them to this theory in order to determine the degree of this correspondence or lack of correspondence in reflecting the meaning of exaggeration in the word. Arabic with its translation in Shahidi and Foladvand translations should be identified as two translations that have a special place in terms of literature. After the investigations, it was observed that these translations could not accurately reflect the meaning of exaggeration in the Arabic language based on this theory and failed to reflect the meaning of exaggeration in these weights.

Regina Solová

“FRIENDS OF POLISH BOOKS” AND “SERVANTS OF EXCELLENCE”: TRANSLATORS’ PROFILES IN THE MONTHLY POLAND (1970-1981) The paper deals with the image of translators in the monthly Poland in the years 1970-1981. Forty articles are analysed with the intention of testing the hypothesis that the promotion of the profiles of translators in the journal was part of a broader campaign launched by the authorities of the Polish People’s Republic to promote the country abroad. The publications in which translators were called “friends of Polish books” and “servants of excellence” were to build a positive image of Poland as a country with an exemplary foreign cultural policy in which translation played an important role. The question to what extent the image-oriented efforts of translators convinced readers about the cultural policy of the Polish People’s Republic remains open.

Hilário Alencar, Gregório Silva Neto

In this paper we prove that two-dimensional translating solitons in $\mathbb{R}^3$ with finite $L$-index are homeomorphic to a plane or a cylinder and that a two-dimensional self-expander with finite $L$-index and sub exponential weighted volume growth has finite topology. We also prove that translating solitons and self-expanders have finite topology, provided the bottom of the spectrum of the $L$-stability operator is bounded from below and their weighted volume have subexponential growth.

Chen Qiu, Dan Oneata, Emanuele Bugliarello et al.

Most vision-and-language pretraining research focuses on English tasks. However, the creation of multilingual multimodal evaluation datasets (e.g. Multi30K, xGQA, XVNLI, and MaRVL) poses a new challenge in finding high-quality training data that is both multilingual and multimodal. In this paper, we investigate whether machine translating English multimodal data can be an effective proxy for the lack of readily available multilingual data. We call this framework TD-MML: Translated Data for Multilingual Multimodal Learning, and it can be applied to any multimodal dataset and model. We apply it to both pretraining and fine-tuning data with a state-of-the-art model. In order to prevent models from learning from low-quality translated text, we propose two metrics for automatically removing such translations from the resulting datasets. In experiments on five tasks across 20 languages in the IGLUE benchmark, we show that translated data can provide a useful signal for multilingual multimodal learning, both at pretraining and fine-tuning.

Nizar El Idrissi, Samir Kabbaj, Brahim Moalige

If $H$ is a Hilbert space, the Stiefel manifold $St(n,H)$ is formed by all the independent $n$-tuples in $H$. In this article, we contribute to the topological study of Stiefel manifolds by proving a path-connectedness result. We prove that the intersection of translates of $St(n,H)$ is path-connected by polygonal paths under a condition on the codimension of the span of the components of the translating $n$-tuples. We rely on a lemma that we prove for the occasion.

Elena Ruiz-Cortés

Prateek Verma, Constantin Basica, Pamela Davis Kivelson

We propose a system that learns from artistic pairings of music and corresponding album cover art. The goal is to 'translate' paintings into music and, in further stages of development, the converse. We aim to deploy this system as an artistic tool for real time 'translations' between musicians and painters. The system's outputs serve as elements to be employed in a joint live performance of music and painting, or as generative material to be used by the artists as inspiration for their improvisation.

Huda Khayrallah, Brian Thompson, Matt Post et al.

Many valid translations exist for a given sentence, yet machine translation (MT) is trained with a single reference translation, exacerbating data sparsity in low-resource settings. We introduce Simulated Multiple Reference Training (SMRT), a novel MT training method that approximates the full space of possible translations by sampling a paraphrase of the reference sentence from a paraphraser and training the MT model to predict the paraphraser's distribution over possible tokens. We demonstrate the effectiveness of SMRT in low-resource settings when translating to English, with improvements of 1.2 to 7.0 BLEU. We also find SMRT is complementary to back-translation.