Hasil untuk "English literature"

Aidana Aidynkyzy, Oğuz Dikenelli, Oylum Alatlı et al.

The scarcity of annotated datasets for clinical information extraction in non-English languages hinders the evaluation of large language model (LLM)-based methods developed primarily in English. In this study, we present the first comprehensive bilingual evaluation of LLMs for the clinical Relation Extraction (RE) task in both English and Turkish. To facilitate this evaluation, we introduce the first English-Turkish parallel clinical RE dataset, derived and carefully curated from the 2010 i2b2/VA relation classification corpus. We systematically assess a diverse set of prompting strategies, including multiple in-context learning (ICL) and Chain-of-Thought (CoT) approaches, and compare their performance to fine-tuned baselines such as PURE. Furthermore, we propose Relation-Aware Retrieval (RAR), a novel in-context example selection method based on contrastive learning, that is specifically designed to capture both sentence-level and relation-level semantics. Our results show that prompting-based LLM approaches consistently outperform traditional fine-tuned models. Moreover, evaluations for English performed better than their Turkish counterparts across all evaluated LLMs and prompting techniques. Among ICL methods, RAR achieves the highest performance, with Gemini 1.5 Flash reaching a micro-F1 score of 0.906 in English and 0.888 in Turkish. Performance further improves to 0.918 F1 in English when RAR is combined with a structured reasoning prompt using the DeepSeek-V3 model. These findings highlight the importance of high-quality demonstration retrieval and underscore the potential of advanced retrieval and prompting techniques to bridge resource gaps in clinical natural language processing.

Sen Yang, Yu Bao, Yu Lu et al.

Large language models (LLMs) have demonstrated strong machine translation capabilities for English-centric language pairs but underperform in direct non-English (x2x) translation. This work addresses this limitation through a synthetic data generation framework that leverages models' established English-to-x (en2x) capabilities. By extending English parallel corpora into omnidirectional datasets and developing an English-referenced quality evaluation proxy, we enable effective collection of high-quality x2x training data. Combined with preference-based optimization, our method achieves significant improvement across 72 x2x directions for widely used LLMs, while generalizing to enhance en2x performance. The results demonstrate that strategic exploitation of English-centric strengths can bootstrap comprehensive multilingual translation capabilities in LLMs. We release codes, datasets, and model checkpoints at https://github.com/NJUNLP/EAX

Michael Hoffmann, Jophin John, Stefan Schweter et al.

We present Llama-GENBA-10B, a trilingual foundation model addressing English-centric bias in large language models. Built on Llama 3.1-8B and scaled to 10B parameters, Llama-GENBA-10B is continuously pretrained on 164B tokens (82B English, 82B German, and 80M Bavarian), balancing resources while preventing English dominance. Targeted at the German NLP community, the model also promotes Bavarian as a low-resource language. Development tackled four challenges: (1) curating a multilingual corpus despite Bavarian scarcity, (2) creating a unified tokenizer for English, German, and Bavarian, (3) optimizing architecture and language-ratio hyperparameters for cross-lingual transfer, and (4) establishing the first standardized trilingual evaluation suite by translating German benchmarks into Bavarian. Evaluations show that Llama-GENBA-10B achieves strong cross-lingual performance, with the fine-tuned variant surpassing Apertus-8B-2509 and gemma-2-9b in Bavarian and establishing itself as the best model in its class for this language, while also outperforming EuroLLM in English and matching its results in German. Training on the Cerebras CS-2 demonstrated efficient large-scale multilingual pretraining with documented energy use, offering a blueprint for inclusive foundation models that integrate low-resource languages.

Zining Wang, Yuxuan Zhang, Dongwook Yoon et al.

With more than 11 times as many pageviews as the next largest edition, English Wikipedia dominates global knowledge access relative to other language editions. Readers are prone to assuming English Wikipedia as a superset of all language editions, leading many to prefer it even when their primary language is not English. Other language editions, however, comprise complementary facts rooted in their respective cultures and media environments, which are marginalized in English Wikipedia. While Wikipedia's user interface enables switching between language editions through its Interlanguage Link (ILL) system, it does not reveal to readers that other language editions contain valuable, complementary information. We present WikiGap, a system that surfaces complementary facts sourced from other Wikipedias within the English Wikipedia interface. Specifically, by combining a recent multilingual information-gap discovery method with a user-centered design, WikiGap enables access to complementary information from French, Russian, and Chinese Wikipedia. In a mixed-methods study (n=21), WikiGap significantly improved fact-finding accuracy, reduced task time, and received a 32-point higher usability score relative to Wikipedia's current ILL-based navigation system. Participants reported increased awareness of the availability of complementary information in non-English editions and reconsidered the completeness of English Wikipedia. WikiGap thus paves the way for improved epistemic equity across language editions.

Maryam Vafapour, Hanieh Talebi, Mahsa Danaei et al.

Abstract Background This study investigates the relationship between polymorphisms in the MTHFR gene and the risk of preterm birth (PTB). Methods A comprehensive literature review was conducted using databases such as PubMed, Web of Science, and CNKI, with the search finalized on January 1, 2025. The review specifically targeted studies published prior to this date, utilizing relevant keywords and MeSH terms associated with PTB and genetic factors. Inclusion criteria encompassed original case-control, longitudinal, or cohort studies, with no limitations on language or publication date. Associations were quantified using odds ratios (ORs) and 95% confidence intervals (CIs) via Comprehensive Meta-Analysis software. Results The analysis included 44 case-control studies comprising 7,384 cases and 51,449 controls, extracted from 28 publications in both English and Chinese. Among these studies, 29 focused on the MTHFR C677T polymorphism, while 15 examined the MTHFR A1298C variant. Pooled results demonstrated a significant association between the MTHFR C677T polymorphism and PTB under five genetic models: allele (C vs. T; OR = 1.303, 95% CI 1.151–1.475, p ≤ 0.001), homozygote (CC vs. AA; OR = 1.494, 95% CI 1.212–1.842, p ≤ 0.001), heterozygote (CT vs. AA; OR = 1.303, 95% CI 1.119–1.516, p = 0.001), dominant (CC + CT vs. AA; OR = 1.341, 95% CI 1.161–1.548, p ≤ 0.001), and recessive (CC vs. CT + AA; OR = 1.340, 95% CI 1.119–1.604, p = 0.001). Subgroup analyses indicated significant associations in Asian populations, particularly in studies conducted in China and India, while no significant correlations were found in Caucasian populations, including those from Austria. Moreover, the MTHFR A1298C polymorphism did not demonstrate a significant relationship with PTB risk across the studied ethnicities. Conclusions The findings indicate a significant association between the MTHFR C677T polymorphism and PTB risk, particularly in Asian and Indian populations, while no significant associations were identified in Caucasian groups. Conversely, the MTHFR A1298C polymorphism appeared to have a negligible impact on PTB risk, underscoring the importance of considering population-specific factors in understanding the genetic epidemiology of PTB.

Monika Zabrocka

This article explores the dynamic evolution of audiovisual translation (AVT) and media accessibility amid technological advances and shifting societal expectations. Once viewed as simple linguistic aids, AVT and media accessibility now play a vital role in inclusive cultural participation and global communication. The text highlights diverse practices – from subtitling and dubbing to game localisation and services such as audio description and closed captions – showcasing their creative, user-focused transformations. It also serves as an introduction, previewing the contributions in this issue that collectively reimagine AVT and accessibility as interconnected, evolving fields essential for fostering equitable, immersive, and culturally sensitive media experiences worldwide.

Saubhagya Joshi, Eunbin Ha, Yonaira Rivera et al.

ChatGPT is a popular information system (over 1 billion visits in August 2023) that can generate natural language responses to user queries. It is important to study the quality and equity of its responses on health-related topics, such as vaccination, as they may influence public health decision-making. We use the Vaccine Hesitancy Scale (VHS) proposed by Shapiro et al.1 to measure the hesitancy of ChatGPT responses in English, Spanish, and French. We find that: (a) ChatGPT responses indicate less hesitancy than those reported for human respondents in past literature; (b) ChatGPT responses vary significantly across languages, with English responses being the most hesitant on average and Spanish being the least; (c) ChatGPT responses are largely consistent across different model parameters but show some variations across the scale factors (vaccine competency, risk). Results have implications for researchers interested in evaluating and improving the quality and equity of health-related web information.

Dipankar Srirag, Jordan Painter, Aditya Joshi et al.

Existing benchmarks often fail to account for linguistic diversity, like language variants of English. In this paper, we share our experiences from our ongoing project of building a sentiment classification benchmark for three variants of English: Australian (en-AU), Indian (en-IN), and British (en-UK) English. Using Google Places reviews, we explore the effects of various sampling techniques based on label semantics, review length, and sentiment proportion and report performances on three fine-tuned BERT-based models. Our initial evaluation reveals significant performance variations influenced by sample characteristics, label semantics, and language variety, highlighting the need for nuanced benchmark design. We offer actionable insights for researchers to create robust benchmarks, emphasising the importance of diverse sampling, careful label definition, and comprehensive evaluation across linguistic varieties.

Antón de la Fuente, Dan Jurafsky

This study asks how self-supervised speech models represent suprasegmental categories like Mandarin lexical tone, English lexical stress, and English phrasal accents. Through a series of probing tasks, we make layer-wise comparisons of English and Mandarin 12 layer monolingual models. Our findings suggest that 1) English and Mandarin wav2vec 2.0 models learn contextual representations of abstract suprasegmental categories which are strongest in the middle third of the network. 2) Models are better at representing features that exist in the language of their training data, and this difference is driven by enriched context in transformer blocks, not local acoustic representation. 3) Fine-tuned wav2vec 2.0 improves performance in later layers compared to pre-trained models mainly for lexically contrastive features like tone and stress, 4) HuBERT and WavLM learn similar representations to wav2vec 2.0, differing mainly in later layer performance. Our results extend previous understanding of how models represent suprasegmentals and offer new insights into the language-specificity and contextual nature of these representations.

Philip D. Bonomi, Jeffrey Crawford, Richard F. Dunne et al.

Abstract Cachexia, with weight loss (WL) as a major component, is highly prevalent in patients with cancer and indicates a poor prognosis. The primary objective of this study was to conduct a meta‐analysis to estimate the risk of mortality associated with cachexia (using established WL criteria prior to treatment initiation) in patients with non‐small‐cell lung cancer (NSCLC) in studies identified through a systematic literature review. The review was conducted according to PRISMA guidelines. Embase® and PubMed were searched to identify articles on survival outcomes in adult patients with NSCLC (any stage) and cachexia published in English between 1 January 2016 and 10 October 2021. Two independent reviewers screened titles, abstracts and full texts of identified records against predefined inclusion/exclusion criteria. Following a feasibility assessment, a meta‐analysis evaluating the impact of cachexia, defined per the international consensus criteria (ICC), or of pre‐treatment WL ≥ 5% without a specified time interval, on overall survival in patients with NSCLC was conducted using a random‐effects model that included the identified studies as the base case. The impact of heterogeneity was evaluated through sensitivity and subgroup analyses. The standard measures of statistical heterogeneity were calculated. Of the 40 NSCLC publications identified in the review, 20 studies that used the ICC for cachexia or reported WL ≥ 5% and that performed multivariate analyses with hazard ratios (HRs) or Kaplan–Meier curves were included in the feasibility assessment. Of these, 16 studies (80%; n = 6225 patients; published 2016–2021) met the criteria for inclusion in the meta‐analysis: 11 studies (69%) used the ICC and 5 studies (31%) used WL ≥ 5%. Combined criteria (ICC plus WL ≥ 5%) were associated with an 82% higher mortality risk versus no cachexia or WL < 5% (pooled HR [95% confidence interval, CI]: 1.82 [1.47, 2.25]). Although statistical heterogeneity was high (I2 = 88%), individual study HRs were directionally aligned with the pooled estimate, and there was considerable overlap in CIs across included studies. A subgroup analysis of studies using the ICC (HR [95% CI]: 2.26 [1.80, 2.83]) or WL ≥ 5% (HR [95% CI]: 1.28 [1.12, 1.46]) showed consistent findings. Assessments of methodological, clinical and statistical heterogeneity indicated that the meta‐analysis was robust. Overall, this analysis found that ICC‐defined cachexia or WL ≥ 5% was associated with inferior survival in patients with NSCLC. Routine assessment of both weight and weight changes in the oncology clinic may help identify patients with NSCLC at risk for worse survival, better inform clinical decision‐making and assess eligibility for cachexia clinical trials.

Stephanie D. Jamieson, Michelle R. Tuckey

Richard Kimera, Daniela N. Rim, Heeyoul Choi

Neural machine translation (NMT) has achieved great successes with large datasets, so NMT is more premised on high-resource languages. This continuously underpins the low resource languages such as Luganda due to the lack of high-quality parallel corpora, so even 'Google translate' does not serve Luganda at the time of this writing. In this paper, we build a parallel corpus with 41,070 pairwise sentences for Luganda and English which is based on three different open-sourced corpora. Then, we train NMT models with hyper-parameter search on the dataset. Experiments gave us a BLEU score of 21.28 from Luganda to English and 17.47 from English to Luganda. Some translation examples show high quality of the translation. We believe that our model is the first Luganda-English NMT model. The bilingual dataset we built will be available to the public.

Kamran Akhtar Siddiqui, Rashid Hussain Abbasi, Ikramullah Khan et al.

Reading and listening are two receptive skills indispensable for English language acquisition. Students are given reading and listening comprehension tests in academic settings, increasing the importance of these skills. However, reading and listening comprehension depends on various factors, including learning styles. While previous research in Pakistan has explored the connection between learning styles and academic success, there is a gap in the literature in understanding how specific learning styles, such as visual and auditory, influence fresh undergraduates' reading and listening comprehension. Therefore, employing a causal-comparative design under a quantitative paradigm, this study examined whether reading and listening comprehension differed based on their students' learning styles (visual and auditory). The data was collected using a learning style questionnaire, a reading comprehension test, and a listening comprehension test, and it was analyzed using an independent sample t-test. The findings revealed that students' reading and listening comprehension scores did not significantly differ based on their preferred learning styles, visual or auditory. The findings indicated that sig. values of reading comprehension (0.672) and listening comprehension (0.668) were greater than the p-value of 0.05. Hence, H0 was accepted, and Ha was rejected. The study will have implications for teachers who want to adapt their teaching methods to fit the individual learning styles of their students.

Trisnowati Tanto, Jeanyfer Tanusy

J.K. Rowling and her Harry Potter series, published from 1997 to 2007, have been globally acclaimed as one of the most popular novels with the most varied target readers. A lot of reviews have been made and by far, they focused more on the literary aspects. This research offers a different way of reading Harry Potter novels since it will focus more on language play as Rowling’s style of writing, which is believed to contribute to the comprehension of the literary elements. Language play is a means of foregrounding – linguistic forms that stand out in a text. The research is done to find out the language play forms contained in Harry Potter and the Philosopher’s Stone and how these forms support the understanding of the literary elements. The research employs the descriptive qualitative method; various linguistic forms of language play are taken as the data, which are then investigated further in terms of the types of linguistic features. Afterward, these are connected with the literal and contextual meanings so as to find out how they support the literary elements. The findings show that besides being entertaining, the language play helps support reader’s imagination as well as supporting the characterization and plot. The research also shows that language play deserves considerable appreciation since creating as well as understanding it needs a high language skill from both the author and readers, especially when it is intended not only to be entertaining but to support the literary elements.

Chenyang Meng, Boyong Jiang, Ming Liu et al.

BackgroundRotator cuff injuries are common, and morbidity increases with age. The asymptomatic full-thickness tear rate is 40% in the over 75-year-old population.PurposeThis study aimed to systematically review the literature on the outcomes of rotator cuff repair among &gt;75 years old patients.Study designSystematic review.MethodsA systematic review of the literature was performed following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A literature search was performed in the electronic databases of PubMed, Medline, Embase, and The Cochrane Library. Studies in English evaluating repair of full-thickness rotator cuff tears in patients aged &gt;75 years were included.ResultsSix studies were reviewed, including 311 patients (313 shoulders) treated with arthroscopic and/or open rotator cuff repair. Sixty-one patients were lost to follow-up, leaving 252 shoulders with outcome data. Patients in this age group demonstrated a significant improvement in the clinical and functional scores after rotator cuff repair, with a high satisfaction rate. The mean American Shoulder and Elbow Surgeons scores improved from 43.8 (range, 42.0–45.5) preoperatively to 85.3 (range, 84.0 to 86.5) postoperatively, and the mean Constant scores improved from 45.4 (range, 34.7–55.5) to 78.6 (range, 67.0–91.6). Pain, evaluated in all studies by the visual analog scale for pain, showed a significant improvement at the last follow-up compared with the mean preoperative score. Furthermore, range of motion and return to daily activities and sports gained marked improvements.ConclusionRotator cuff repair in patients aged &gt;75 years could achieve high clinical success rates with good outcomes and pain relief. Although patients in this age group are at a high risk of retear, rotator cuff repair may offer a good option with significant functional and clinical improvement.

Irina A. Barkova, Margarita P. Izhberdeeva, Anastasiya A. Sautkina

Bacteriophage endolysins are a biologically active substances that play a specific role in the release of phage progeny by degrading the peptidoglycan of the host bacterium. In the light of antibiotic resistance, endolysins are considered as alternative therapeutic agents because of their exceptional ability to target bacterial cells. Aim summarization of the data on the biology, structure, mechanisms of action of bacteriophage endolysins, as well as on preparations based on them, which are at different stages of research. The results of studies of bacterial endolysins over the past 20 years were searched using the Internet resources PubMed, Web of Science, Scopus in English for the keywords: lysin, bacteriophages, holin, antibiotic resistance. The analysis of literature data showed that the structure of phage endolysins of Gram-positive and Gram-negative bacteria differs from each other and reflects differences in their architecture due to variation in the cell wall composition of these two major bacterial groups. Depending on the cleavable bond in peptidoglycan, endolysins can be divided into at least five different groups: glycosidases (two groups aminidases and muramidases), endopeptidases, specific amidogyrolases, and lytic transglycosylases. To date, endolysins effective against a number of pathogens have been studied, including Streptococcus pneumoniae, Streptococcus pyogenes, Streptococcus agalactiae, Staphylococcus aureus, Mycobacterium spp., Pseudomonas aeruginosa, etc. A number of studies have shown the therapeutic potential of endolysins in combating antibiotic-resistant infections.

Roi Boy Jon, R. Embong, B. Purnama et al.

The widespread use of English worldwide has brought about a significant impact for human beings to date. In the education sphere mainly, many studies have been conducted to discover the issues in English instruction. Moreover, English teachers in Indonesia were highly encouraged to figure out the best method to teach and cope with any problems encountered by the students to achieve the best learning outcomes. Besides, the government had also tried its best endeavors to accustom the curriculum to the needs in this globalization era. However, the results have not always been as expected since teachers and students possessed divergent points of view linked with the implementation of English language instruction. Therefore, this scientific literature review discussed some issues related to English which included; English in Indonesia, English Language Teaching for Education, Teachers’ Strategies to English Language Teaching, Problems of English Language Teaching, and Students’ Perceptions towards English Language Teaching. Furthermore, the main aim of this paper was to acknowledge to the readers that the teachers and students essentially require English due to its crucial role in the development of Education in Indonesia.

Nada Bela Andresta, Choiril Anwar

This study aims to determine students' perceptions of using Quizizz as an English assessment medium. This study uses a descriptive qualitative method taken one month to collect data using close-ended questionnaire as the instrument. The sample used in this study was 3 classes in Grade 10 at SMA N 2 Semarang in the 2021/2022 academic year with a total of 108 students. From the data collection of the questionnaire distribution, it shows that 88.9% or 96 of students have positive perceptions and 5.6% or 6 of students have neutral perceptions, and 5,6% or 6 students have negative perceptions of the use of Quizizz as a medium for assessing English in their classes. It can be concluded that students in Grade 10 at SMA N 2 Semarang have a positive perception of the use of Quizizz. The students thought that the Quizizz media was a fun medium and very suitable to be used by teachers because the media was very interesting and practical to use for English assessment media.

Habil Otanga, Brian Semujju, Lynn Mwaniki et al.

<h4>Background</h4> Diabetes mellitus is a growing worldwide health challenge especially in sub-Saharan Africa. While the use and effectiveness of diabetes self-management interventions is well documented in high-income countries, little information exists in sub-Saharan Africa. Therefore, this study attempted to synthesize information in the literature on the use and efficacy of peer support and social networking in diabetes self-management in Kenya and Uganda. <h4>Objective</h4> The purpose of this scoping review is to summarize research on the extent of use and efficacy of peer support and social networking interventions in diabetes self-management in Kenya and Uganda. <h4>Design</h4> We searched PubMed, ScienceDirect and Cochrane Library databases for articles reporting peer support and social networking interventions in Kenya and Uganda published in English between 2000 and September 2021. Key words encapsulated three major themes: peer support, social networking and self-management. Hand searches were also conducted to select eligible papers. Data was extracted using a form prepared and piloted in line with PRISMA-ScR guidelines. <h4>Results</h4> Thirteen peer reviewed articles were selected for analysis. Eleven studies reported peer support interventions while two focused on social networks in diabetes self-management. Peer support and social networking interventions incorporated microfinance and group medical visits, diabetes self-management education, telephone support and Medication Adherence Clubs. Most interventions were delivered by multidisciplinary teams comprising nurses and other professionals, peer educators, peer leaders and community health workers. Most interventions were effective and led to improvements in HbA1c and blood pressure, eating behaviors and physical activity and social support. <h4>Conclusions</h4> The limited studies available show that peer support and social networking interventions have mixed results on health and other outcomes. Importantly, most studies reported significant improvements in clinical outcomes. Further research is needed on the nature and mechanisms through which peer support and social network characteristics affect health outcomes.

Luke Laari, Oboshie Anim-Boamah, Christian Makafui Boso

Background & Aim: While hard skills refer to the technical ability and factual knowledge needed to do a job, soft skills allow you to use your technical abilities and knowledge more effectively. These two skills are complementary, but soft skills are prerequisites in every profession where human interaction and teamwork are needed to succeed. This integrative review examined the research on soft skills in nursing and made recommendations based on its findings. Methods & Materials: Whittemore and Knafl’s five-step integrative review framework was carried out using four electronic databases. These databases are the Cumulative Index to Nursing & Allied Health Literature (CINAHL), PubMed, Medline on EBSCOhost, and Scopus. Searches were conducted using keywords: soft skills, non-technical skills, nursing skills, nursing art, and aesthetics. The literature search explored no date ranges, and only the English language was considered. Full texts of relevant studies in both qualitative and quantitative research were retrieved. Critical appraisal was undertaken, and the findings of the relevant studies were analyzed using thematic analysis. Results: Seventeen studies were included, and the findings suggest an urgent need for soft skills in the nursing domain. Five themes emerged: the meaning of soft skills in nursing, the benefits of soft skills in nursing; the need for soft skills in nursing; the incorporation of soft skills into nursing practice; and the relationship between hard and soft skills. Findings show soft skills are the cognitive and social capabilities that complete the technical skills of the nurse. Conclusion: Incorporating soft skills into the nursing curriculum should be a resuscitative call that requires immediate attention.