Hasil untuk "English language"

Albert Maganaka

This critical autoethnography examines how intersecting identities—race, gender expression, linguistic background, and migration—shape professional legitimacy in English language education. Drawing on counter‑storytelling within a Critical Race Theory and intersectionality framework, the study centers the lived experience of a Filipino‑Canadian non‑native English‑speaking teacher (NNEST) and language assessor. Narratives trace early gendered exclusion, the internalization and critique of native‑speakerism, and a shift toward intelligibility‑focused assessment and translanguaging‑informed pedagogy. The analysis demonstrates how legitimacy is negotiated across individual encounters and institutional structures, and how reflective practice transforms deficit framings into asset‑based, identity‑affirming instruction. Findings highlight three implications: (1) redefining legitimacy beyond “nativeness” to emphasize pedagogical competence and lived expertise; (2) positioning identity negotiation and resilience as central to educator development; and (3) using counter‑storytelling to challenge dominant ideologies and inform inclusive assessment. Recommendations address teacher education, mentorship, anti‑bias admissions and hiring, and policies that value multilingualism and protect gender identity and expression.

Tim Noble

This article focuses on the “terrible sonnets” of the English poet Gerard Manley Hopkins (1844-1889). It shows how these poems offer an insight into a particular form of agonistics. Hopkins struggles with his understanding of self, of God, and of language, transforming experience into a fundamental clash of hegemonic visions of the world. Through resilience, there is not reconciliation but a re-visioning (therefore remediatising) of the possibility of human existence, which finds its home in language. Hopkins thus works between sound and sight and places a spiritual dimension into the political conversation about the task of being human. This challenge to any reductive approach is also, the article argues, an expression of resilience.

Musthamin Balumbi, Stang Stang, Suriah Suriah et al.

Reproductive health education at the primary school level is a controversial topic. Although some recognize its importance in providing accurate information about the body and development, others raise concerns about cultural sensitivities and age-appropriateness. This review intends to explore various aspects related to the importance of reproductive health education among primary school children, as well as the challenges and benefits associated. This article presents a literature review of previous studies on the implementation of health and reproductive education in elementary school children. An extensive search was conducted to identify relevant papers using relevant databases like ScienceDirect, PubMed, and Google Scholar. The articles included were selected if published between 2013 and 2023, in the English language, and have undergone a rigorous peer-review process. Our review identified substantial benefits of reproductive health education in primary schools. Studies showed a positive impact on reducing misconceptions about reproduction, promoting healthy attitudes towards the body, and potentially lowering risks of teenage pregnancy and sexually transmitted diseases. However, the review also revealed significant challenges. Cultural and religious sensitivities often lead to resistance from some communities. Additionally, ensuring age-appropriate language, content, and delivery methods remains a concern. The findings highlight the need for a balanced approach to reproductive health education in primary schools. While acknowledging cultural sensitivities, strategies like involving communities and using inclusive language can promote inclusivity. Open communication within families and well-trained teachers are crucial for effective reproductive health implementation. By addressing these challenges through inclusive and age-appropriate methods, reproductive health education programs can equip children with the knowledge and skills necessary for a healthy future.

Hankiz Yilahun, Hangtian Zhao, Askar Hamdulla

Few-shot relation extraction (RE) is a critical task in natural language processing (NLP), yet benchmark datasets for non-English languages, particularly Chinese, remain scarce. Owing to their narrow focus and insufficient coverage of relation types, extant Chinese datasets typically exhibit limited applicability and it proves difficult to establish a standardized evaluation benchmark. To address this issue, we introduce FREDC, a few-shot RE dataset for Chinese (FREDC) featuring nearly 100 relation categories across general, medical, and financial domains, with over 100,000 annotated instances. FREDC supports diverse experimental setups, including cross-domain and “None-of-the-Above” (NOTA) tasks. To evaluate FREDC, we adapted the baseline RE model designed for English to Chinese and tested across multiple experimental configurations. FREDC has proven to be effective across multiple models. It also posed challenges to cross-domain and NOTA tasks, due to the linguistic complexity and domain-specific features of the dataset. Comparisons with English datasets showed a performance gap of 15–30% on the accuracy. FREDC fills a critical gap by providing a standardized, high-quality resource for Chinese few-shot RE. It offers a robust foundation for advancing multilingual RE research and improving the adaptability of RE models across languages and domains. We provide relevant code snippets to facilitate reproducibility and encourage further research in the field.

Damian Heywood, Joseph Andrew Carrier, Kyu-Hong Hwang

This study describes the development of an AI-assisted error analysis system designed to identify, categorize, and correct writing errors in English. Utilizing Large Language Models (LLMs) like Claude 3.5 Sonnet and DeepSeek R1, the system employs a detailed taxonomy grounded in linguistic theories from Corder (1967), Richards (1971), and James (1998). Errors are classified at both word and sentence levels, covering spelling, grammar, and punctuation. Implemented through Python-coded API calls, the system provides granular feedback beyond traditional rubric-based assessments. Initial testing on isolated errors refined the taxonomy, addressing challenges like overlapping categories. Final testing used "English as she is spoke" by Jose da Fonseca (1855), a text rich with authentic linguistic errors, to evaluate the system's capacity for handling complex, multi-layered analysis. The AI successfully identified diverse error types but showed limitations in contextual understanding and occasionally generated new error categories when encountering uncoded errors. This research demonstrates AI's potential to transform EFL instruction by automating detailed error analysis and feedback. While promising, further development is needed to improve contextual accuracy and expand the taxonomy to stylistic and discourse-level errors.

Mingzhe Xing, Chang Tian, Jianan Zhang et al.

Modern large-scale networks introduce significant complexity in understanding network behaviors, increasing the risk of misconfiguration. Prior work proposed to understand network behaviors by mining network configurations, typically relying on domain-specific languages interfaced with formal models. While effective, they suffer from a steep learning curve and limited flexibility. In contrast, natural language (NL) offers a more accessible and interpretable interface, motivating recent research on NL-guided network behavior understanding. Recent advances in large language models (LLMs) further enhance this direction, leveraging their extensive prior knowledge of network concepts and strong reasoning capabilities. However, three key challenges remain: 1) numerous router devices with lengthy configuration files challenge LLM's long-context understanding ability; 2) heterogeneity across devices and protocols impedes scalability; and 3) complex network topologies and protocols demand advanced reasoning abilities beyond the current capabilities of LLMs. To tackle the above challenges, we propose NetMind, a novel framework for querying networks using NL. Our approach introduces a tree-based configuration chunking strategy to preserve semantic coherence while enabling efficient partitioning. We then construct a unified fact graph as an intermediate representation to normalize vendor-specific configurations. Finally, we design a hybrid imperative-declarative language to reduce the reasoning burden on LLMs and enhance precision. We contribute a benchmark consisting of NL question-answer pairs paired with network configurations. Experiments demonstrate that NetMind achieves accurate and scalable network behavior understanding, outperforming existing baselines.

Yiran Rex Ma

Based on comparable English-Chinese news corpora annotated by Large Language Model (LLM), this paper attempts to explore the differences in constituent order of English-Chinese news from the perspective of functional chunks with adverbial roles, and analyze their typical positional preferences and distribution patterns. It is found that: (1) English news prefers linear narrative of core information first, and functional chunks are mostly post-positioned, while Chinese news prefers overall presentation mode of background first, and functional chunks are often pre-positioned; (2) In SVO structure, both English and Chinese news show differences in the distribution of functional chunks, but the tendency of Chinese pre-positioning is more significant, while that of English post-positioning is relatively mild; (3) When function blocks are co-occurring, both English and Chinese news show high flexibility, and the order adjustment is driven by information and pragmatic purposes. The study reveals that word order has both systematic preference and dynamic adaptability, providing new empirical support for contrastive study of English-Chinese information structure.

Yehua Wang, Yehua Wang, Xumeng Yan et al.

BackgroundThe rising prevalence of herpes simplex type 2 (HSV-2) infection poses a growing global public health challenge. A comprehensive understanding of its epidemiology and burden disparities in China is crucial for informing targeted and effective intervention strategies in the future.MethodsWe followed Cochrane and PRISMA guidelines for a systematic review and included publications published in Chinese and English bibliographic systems until March 31st, 2024. We synthesized HSV-2 seroprevalence data across different population types. We used random-effects models for meta-analyses and conducted meta-regression to assess the association between population characteristics and seroprevalence.ResultsOverall, 23,999 articles were identified, and 402 publications (1,203,362 participants) that reported the overall seroprevalence rates (858 stratified measures) were included. Pooled HSV-2 seroprevalence among the general population (lower risk) was 7.7% (95% CI: 6.8-8.7%). Compared to the general population, there is a higher risk of HSV-2 prevalence among intermediate-risk populations (14.8%, 95% CI: 11.0-19.1%), and key populations (31.7%, 95% CI: 27.4-36.1%). Female sexual workers (FSWs) have the highest HSV-2 risk (ARR:1.69, 95% CI: 1.61-1.78). We found northeastern regions had a higher HSV-2 seroprevalence than other regions (17.0%, 95% CI: 4.3-35.6%, ARR: 1.38, 95% CI: 1.26-1.50, Northern China as the reference group). This highlighted the disparity by population risk levels and regions. We also found lower HSV-2 prevalence estimates in publications in Chinese bibliographic databases than those in English databases among key populations (such as MSM and HIV-discordant populations).ConclusionThere is a gradient increase in HSV-2 prevalence risk stratification. We also identified region, population, and age disparities and heterogeneities by publication language in the HSV-2 burden. This study provides guidance for future HSV-2 prevention to eliminate disparities of HSV-2 infection and reduce overall HSV-2 burden.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/display_record.php?RecordID=408108, identifier CRD42023408108.

Ava Oliaei, Dean Elterman, Salar Sadri et al.

Background and aimFecal incontinence (FI) is defined as the unintended loss of solid or liquid stool. FI adversely affects the patient’s quality of life. However, due to stigma, lack of awareness, and underdiagnosis, there is a notable gap in the knowledge regarding its prevalence. This study aimed to conduct a systematic review and meta-analysis of published literature reporting on FI prevalence and estimate the number of people afflicted by FI.MethodsA systematic review was conducted following the PRISMA 2020 guidelines, using the Embase, MEDLINE, CINHAL, and PubMed databases to identify relevant publications in the English language. Two reviewers independently screened the articles and extracted data. The reference sections and content of the review papers were also evaluated. Thirty-two articles were selected and included. A meta-analysis of proportions was performed using RStudio software. A sub-analysis was conducted to account for the variation between sample population age groups to minimize heterogeneity. The pooled prevalence was extrapolated to the Canadian population and a sample of ten densely populated countries to estimate the number of people affected by FI.ResultsThe Mean pooled FI prevalence in men and women was 7% (95% CI: 6-9%) and 10% (95% CI: 8-12%), respectively. The sub-analysis mean pooled prevalence of FI in men and women was 8% (95% CI: 6-10%) and 10% (95% CI: 8-12%), respectively. The authors estimate that between 1 and 1.5 million Canadians and 320 to 500 million people in the ten most populous countries suffer from FI.ConclusionFecal incontinence is a prevalent underdiagnosed condition requiring appropriate and timely treatment to improve a patient’s quality of life.

Nezha BADI

Abstract: The present study examines the effect of the flow-based-notetaking instruction on Algerian university students’ critical information recalling of the Introduction to General Linguistics (ILG) content. A total number of 76 first year English as a foreign language (EFL) students divided into two groups participate in the study. One group referred to as the experimental group receives the flow-notetaking instruction, whereas, the other group (the control group) is not exposed to any training program. A pre-post-test design and an intervention is implemented to provide answers for the research questions addressed. The data are drawn from a set of ILG content activities and an interview. While the ILG activities are used to elicit the effectiveness of the intervention in enhancing students’ information recalling, the interview is used to scrutinize students’ perceptions of the flow-notetaking intervention. Findings of the study indicate that the intervention that the experimental group undergoes is effective in improving their ILG scores. In other words, although there is no significant difference between the control and the experimental group in their pre-test scores, at post-test, both groups differ significantly, with the experimental group significantly outperforms the control group in their scores. In terms of students’ view about the flow-notetaking intervention, data gathered seem to provide evidence for the effectiveness of the intervention in the students’ eyes, as the overall results reveal positive perceptions. Keywords: EFL students, flow-based-notetaking, information recalling, students’ perceptions.

Olga Iakovenko, Thomas Hain

Code-switching (CS) is the process of speakers interchanging between two or more languages which in the modern world becomes increasingly common. In order to better describe CS speech the Matrix Language Frame (MLF) theory introduces the concept of a Matrix Language, which is the language that provides the grammatical structure for a CS utterance. In this work the MLF theory was used to develop systems for Matrix Language Identity (MLID) determination. The MLID of English/Mandarin and English/Spanish CS text and speech was compared to acoustic language identity (LID), which is a typical way to identify a language in monolingual utterances. MLID predictors from audio show higher correlation with the textual principles than LID in all cases while also outperforming LID in an MLID recognition task based on F1 macro (60%) and correlation score (0.38). This novel approach has identified that non-English languages (Mandarin and Spanish) are preferred over the English language as the ML contrary to the monolingual choice of LID.

Mihai Masala, Denis C. Ilie-Ablachim, Alexandru Dima et al.

In recent years, Large Language Models (LLMs) have achieved almost human-like performance on various tasks. While some LLMs have been trained on multilingual data, most of the training data is in English; hence, their performance in English greatly exceeds other languages. To our knowledge, we are the first to collect and translate a large collection of texts, instructions, and benchmarks and train, evaluate, and release open-source LLMs tailored for Romanian. We evaluate our methods on four different categories, including academic benchmarks, MT-Bench (manually translated), and a professionally built historical, cultural, and social benchmark adapted to Romanian. We argue for the usefulness and high performance of RoLLMs by obtaining state-of-the-art results across the board. We publicly release all resources (i.e., data, training and evaluation code, models) to support and encourage research on Romanian LLMs while concurrently creating a generalizable recipe, adequate for other low or less-resourced languages.

Astrid Johana Aristizábal Cardona, Janeth Maria Ortiz Medina

In a country struggling to emerge from the shadows of a relentless armed conflict, where media manipulation puts a stable reconciliation process at risk, foreign language educators are called on to act. This article presents the results of a study conducted with three foreign language students tudying in one of the sites of a public university in the Department of Antioquia, Colombia, located in a region affected by cultural, structural, direct, and ecological violence. The study resorted to critical peace education and critical media literacy theories as well as to local contributions from a pedagogy of memory to foster critical consciousness about the armed conflict in Colombia. Data collection methods included a survey, recordings of class discussions, samples of students' work, and individual and group interviews. Data showed that the participants reflected critically on the role of different war agents, increased their capacity to question media messages, recognized the relevance of including the victims’ voices, and created counter-texts to contest dominant narratives about the conflict. Findings confirm the urgent need to prepare future foreign language educators to respond to their learners’ harsh social realities and exert their agency to generate transformations. They also highlight the need to do more interdisciplinary work in ELT and to value the contributions of local knowledge that helps us both understand the dynamics of violence in our contexts and envision possibilities for peacebuilding in Colombia.

Hanwool Lee, Jonghyun Choi, Sohyeon Kwon et al.

This paper presents our participation in the FinNLP-2023 shared task on multi-lingual environmental, social, and corporate governance issue identification (ML-ESG). The task's objective is to classify news articles based on the 35 ESG key issues defined by the MSCI ESG rating guidelines. Our approach focuses on the English and French subtasks, employing the CerebrasGPT, OPT, and Pythia models, along with the zero-shot and GPT3Mix Augmentation techniques. We utilize various encoder models, such as RoBERTa, DeBERTa, and FinBERT, subjecting them to knowledge distillation and additional training. Our approach yielded exceptional results, securing the first position in the English text subtask with F1-score 0.69 and the second position in the French text subtask with F1-score 0.78. These outcomes underscore the effectiveness of our methodology in identifying ESG issues in news articles across different languages. Our findings contribute to the exploration of ESG topics and highlight the potential of leveraging advanced language models for ESG issue identification.

Fatmah Al Thobaiti

Newspaper headlines show that awareness of intimate partner violence is a complicated issue that needs further examination. Works of fiction narrated by women trapped in abusive relationships are useful sites for the exploration of what intimate partner violence usually includes, and the identification of subtle behaviours that can be defined as violent and abusive but usually go unnoticed. This article submits two contemporary works of fiction, First Love and the Fifty Shades series, for a study of the covert mechanisms of emotional abuse. To understand such mechanisms, the article engages with feminist as well as postfeminist contemporary thinking on intimate partner violence. The analysis shifts the focus back to the male abuser by carefully depicting how he uses under-recognized, gendered forms of power to abuse his partner. The aim is to elucidate the capacity of first-person narratives to allow access to the abused woman’s mind, while simultaneously provoking questions about the abusers’ behaviours, making them a more powerful tool for understanding intimate partner violence than a newspaper report.

Luca Beurer-Kellner, Marc Fischer, Martin Vechev

Large language models have demonstrated outstanding performance on a wide range of tasks such as question answering and code generation. On a high level, given an input, a language model can be used to automatically complete the sequence in a statistically-likely way. Based on this, users prompt these models with language instructions or examples, to implement a variety of downstream tasks. Advanced prompting methods can even imply interaction between the language model, a user, and external tools such as calculators. However, to obtain state-of-the-art performance or adapt language models for specific tasks, complex task- and model-specific programs have to be implemented, which may still require ad-hoc interaction. Based on this, we present the novel idea of Language Model Programming (LMP). LMP generalizes language model prompting from pure text prompts to an intuitive combination of text prompting and scripting. Additionally, LMP allows constraints to be specified over the language model output. This enables easy adaption to many tasks while abstracting language model internals and providing high-level semantics. To enable LMP, we implement LMQL(short for Language Model Query Language), which leverages the constraints and control flow from an LMP prompt to generate an efficient inference procedure that minimizes the number of expensive calls to the underlying language model. We show that LMQL can capture a wide range of state-of-the-art prompting methods in an intuitive way, especially facilitating interactive flows that are challenging to implement with existing high-level APIs. Our evaluation shows that we retain or increase the accuracy on several downstream tasks, while also significantly reducing the required amount of computation or cost in the case of pay-to-use APIs (26-85% cost savings).

Daniel Korzekwa

Despite significant advances in recent years, the existing Computer-Assisted Pronunciation Training (CAPT) methods detect pronunciation errors with a relatively low accuracy (precision of 60% at 40%-80% recall). This Ph.D. work proposes novel deep learning methods for detecting pronunciation errors in non-native (L2) English speech, outperforming the state-of-the-art method in AUC metric (Area under the Curve) by 41%, i.e., from 0.528 to 0.749. One of the problems with existing CAPT methods is the low availability of annotated mispronounced speech needed for reliable training of pronunciation error detection models. Therefore, the detection of pronunciation errors is reformulated to the task of generating synthetic mispronounced speech. Intuitively, if we could mimic mispronounced speech and produce any amount of training data, detecting pronunciation errors would be more effective. Furthermore, to eliminate the need to align canonical and recognized phonemes, a novel end-to-end multi-task technique to directly detect pronunciation errors was proposed. The pronunciation error detection models have been used at Amazon to automatically detect pronunciation errors in synthetic speech to accelerate the research into new speech synthesis methods. It was demonstrated that the proposed deep learning methods are applicable in the tasks of detecting and reconstructing dysarthric speech.

Afra Feyza Akyürek, Muhammed Yusuf Kocyigit, Sejin Paik et al.

Researchers have devised numerous ways to quantify social biases vested in pretrained language models. As some language models are capable of generating coherent completions given a set of textual prompts, several prompting datasets have been proposed to measure biases between social groups -- posing language generation as a way of identifying biases. In this opinion paper, we analyze how specific choices of prompt sets, metrics, automatic tools and sampling strategies affect bias results. We find out that the practice of measuring biases through text completion is prone to yielding contradicting results under different experiment settings. We additionally provide recommendations for reporting biases in open-ended language generation for a more complete outlook of biases exhibited by a given language model. Code to reproduce the results is released under https://github.com/feyzaakyurek/bias-textgen.

Zhihong Chen, Guanbin Li, Xiang Wan

Medical vision-and-language pre-training (Med-VLP) has received considerable attention owing to its applicability to extracting generic vision-and-language representations from medical images and texts. Most existing methods mainly contain three elements: uni-modal encoders (i.e., a vision encoder and a language encoder), a multi-modal fusion module, and pretext tasks, with few studies considering the importance of medical domain expert knowledge and explicitly exploiting such knowledge to facilitate Med-VLP. Although there exist knowledge-enhanced vision-and-language pre-training (VLP) methods in the general domain, most require off-the-shelf toolkits (e.g., object detectors and scene graph parsers), which are unavailable in the medical domain. In this paper, we propose a systematic and effective approach to enhance Med-VLP by structured medical knowledge from three perspectives. First, considering knowledge can be regarded as the intermediate medium between vision and language, we align the representations of the vision encoder and the language encoder through knowledge. Second, we inject knowledge into the multi-modal fusion model to enable the model to perform reasoning using knowledge as the supplementation of the input image and text. Third, we guide the model to put emphasis on the most critical information in images and texts by designing knowledge-induced pretext tasks. To perform a comprehensive evaluation and facilitate further research, we construct a medical vision-and-language benchmark including three tasks. Experimental results illustrate the effectiveness of our approach, where state-of-the-art performance is achieved on all downstream tasks. Further analyses explore the effects of different components of our approach and various settings of pre-training.

Gerhard Fischer

Remembering Mudrooroo (1938–2019)