The advancing fluency of LLMs raises important questions about their ability to emulate complex human traits, including emotional expression and personality, across diverse linguistic and cultural contexts. This study investigates whether LLMs can convincingly mimic emotional nuance in English and personality markers in Arabic, a critical under-resourced language with unique linguistic and cultural characteristics. We conduct two tasks across six models:Jais, Mistral, LLaMA, GPT-4o, Gemini, and DeepSeek. First, we evaluate whether machine classifiers can reliably distinguish between human-authored and AI-generated texts. Second, we assess the extent to which LLM-generated texts exhibit emotional or personality traits comparable to those of humans. Our results demonstrate that AI-generated texts are distinguishable from human-authored ones (F1>0.95), though classification performance deteriorates on paraphrased samples, indicating a reliance on superficial stylistic cues. Emotion and personality classification experiments reveal significant generalization gaps: classifiers trained on human data perform poorly on AI-generated texts and vice versa, suggesting LLMs encode affective signals differently from humans. Importantly, augmenting training with AI-generated data enhances performance in the Arabic personality classification task, highlighting the potential of synthetic data to address challenges in under-resourced languages. Model-specific analyses show that GPT-4o and Gemini exhibit superior affective coherence. Linguistic and psycholinguistic analyses reveal measurable divergences in tone, authenticity, and textual complexity between human and AI texts. These findings have implications for affective computing, authorship attribution, and responsible AI deployment, particularly within underresourced language contexts where generative AI detection and alignment pose unique challenges.
This research is a forensic linguistic study that concentrates on the analysis of speech acts spoken by a companion from one of the representative offices of a ministry in Cilacap during an investigative interview between an investigator and a child victim of a sexual violence crime. The purpose of this investigation is to examine the speech of a child victim and a companion during an investigative interview that occurred at a police station in Cilacap. This investigation is qualitative in nature and is structured as a case study. Speech data were collected during the investigative interview using listening and note-taking techniques. The data were subsequently analyzed in accordance with Weigand’s (2010) dialogic speech act theory. Additionally, the function and authority of a companion in the investigative interview process are clarified by the Regulation of the Minister of Women Empowerment and Child Protection of the Republic of Indonesia (Permen PPPA) No. 2 of 2022. The analysis results indicate that the code of ethics outlined in Permen PPPA No. 2 of 2022 is contravened by the companion’s dominance of the explorative speech act and the presence of the directive speech act. This implies that the companion must be re-informed about the code of ethics outlined in the Women Empowerment and Child Protection Regulation No. 2 of 2022. The findings of this research have the potential to assist the relevant ministries in enhancing the efficacy and authority of an assistant in the interview process related to the investigation of sexual violence crimes.
Abstract Despite its feasibility and potential, technology-supported peer feedback in foreign language speaking classrooms has been under-researched. The present study aimed to extend this line of research by modeling Chinese postgraduates’ and undergraduates’ motivation, behavioral engagement in feedback provision, and achievement in a particular type of technology-supported peer feedback, i.e., group peer feedback via WeChat in EFL speaking classrooms. The results indicated that the two motivational constructs investigated in the study, namely, expectancy for success and subjective task value did not predict learners’ behavioral engagement in feedback provision; behavioral engagement in feedback provision significantly predicted the assessors’ achievement; and educational level did not moderate the relations between motivation, behavioral engagement in feedback provision, and achievement. The results pointed to the complexity of the relations between motivation and engagement in complicated contexts like technology-supported peer feedback. They also highlighted the value of mobile based group peer feedback tasks in EFL speaking classrooms. Theoretical and pedagogical implications were discussed.
Special aspects of education, Language acquisition
Chesi's (forthcoming) target paper depicts a generative linguistics in crisis, foreboded by Piantadosi's (2023) declaration that "modern language models refute Chomsky's approach to language." In order to survive, Chesi warns, generativists must hold themselves to higher standards of formal and empirical rigor. This response argues that the crisis described by Chesi and Piantadosi actually has little to do with rigor, but is rather a reflection of generativists' limited social ambitions. Chesi ties the fate of generative linguistics to its intellectual merits, but the current success of language model research is social in nature as much as it is intellectual. In order to thrive, then, generativists must do more than heed Chesi's call for rigor; they must also expand their ambitions by giving outsiders a stake in their future success.
Vision Language Navigation (VLN) typically requires agents to navigate to specified objects or remote regions in unknown scenes by obeying linguistic commands. Such tasks require organizing historical visual observations for linguistic grounding, which is critical for long-sequence navigational decisions. However, current agents suffer from overly detailed scene representation and ambiguous vision-language alignment, which weaken their comprehension of navigation-friendly high-level scene priors and easily lead to behaviors that violate linguistic commands. To tackle these issues, we propose a navigation policy by recursively summarizing along-the-way visual perceptions, which are adaptively aligned with commands to enhance linguistic grounding. In particular, by structurally modeling historical trajectories as compact neural grids, several Recursive Visual Imagination (RVI) techniques are proposed to motivate agents to focus on the regularity of visual transitions and semantic scene layouts, instead of dealing with misleading geometric details. Then, an Adaptive Linguistic Grounding (ALG) technique is proposed to align the learned situational memories with different linguistic components purposefully. Such fine-grained semantic matching facilitates the accurate anticipation of navigation actions and progress. Our navigation policy outperforms the state-of-the-art methods on the challenging VLN-CE and ObjectNav tasks, showing the superiority of our RVI and ALG techniques for VLN.
The fusion of language and vision in large vision-language models (LVLMs) has revolutionized deep learning-based object detection by enhancing adaptability, contextual reasoning, and generalization beyond traditional architectures. This in-depth review presents a structured exploration of the state-of-the-art in LVLMs, systematically organized through a three-step research review process. First, we discuss the functioning of vision language models (VLMs) for object detection, describing how these models harness natural language processing (NLP) and computer vision (CV) techniques to revolutionize object detection and localization. We then explain the architectural innovations, training paradigms, and output flexibility of recent LVLMs for object detection, highlighting how they achieve advanced contextual understanding for object detection. The review thoroughly examines the approaches used in integration of visual and textual information, demonstrating the progress made in object detection using VLMs that facilitate more sophisticated object detection and localization strategies. This review presents comprehensive visualizations demonstrating LVLMs' effectiveness in diverse scenarios including localization and segmentation, and then compares their real-time performance, adaptability, and complexity to traditional deep learning systems. Based on the review, its is expected that LVLMs will soon meet or surpass the performance of conventional methods in object detection. The review also identifies a few major limitations of the current LVLM modes, proposes solutions to address those challenges, and presents a clear roadmap for the future advancement in this field. We conclude, based on this study, that the recent advancement in LVLMs have made and will continue to make a transformative impact on object detection and robotic applications in the future.
It is sometimes argued that (certain types of) lexical frequency and constructional frequency determine how easy sentences are to process and hence, how acceptable speakers find them. Others have argued that grammatical principles interact with and often override such effects. Here, we present the results from a survey on Danish with more than 200 participants. We asked people to provide acceptability ratings of a number of sentences with varying levels of complexity, with and without extraction, including complement clauses, relative clauses, parasitic gaps, and ungrammatical sentences. We predicted structural complexity and acceptability to be negatively correlated (the more complex, the less acceptable). The results show that construction frequency and acceptability are correlated, but that zero and near-zero frequencies do not predict acceptability. However, there is indeed an even stronger inverse correlation between acceptability and structural complexity, defined as a function of independently motivated factors of syntactic structure and processing, including embedding, adjunction, extraction, and distance between filler and gap. Lexical frequency also affects acceptability, but the effects are small, and, crucially, there is no evidence in our data that ungrammatical sentences are affected by such frequency effects. Furthermore, the acceptability patterns seem to be fairly stable across participants. The results show a pattern that is consistent with an approach based on grammatical principles and processing constraints, rather than based on stochastic principles alone.
This paper analyses visual discourse in Ahmed Khalid Tawfiq’s The Legend of the Late Night (2002) by investing in semiotics as a critical method, gripping the essence of the image, especially if it involves an imagination imbued with horror literature, as it will give rise to a kind of novelistic narrative through the enjoyment and amazement at the new form that jumps to the mind to formal camouflage with an artistic and functional impact. The prominent approach to tackle this reading is characterized by its ability to analyze literary discourse, reveal its secrets, and interrogate its symbols to reach its goal or approach. Critical reading is considered a simulation analysis of the conceptual procedure through analyzing the physical form of the sign within the discourse or image, studying the semiotic units such as words, colours, shapes and images, and then linking the signs to their cultural, social and other contexts, and an attempt to observe the effect of narrative employment of two types that come together to establish a genre that inspires in terms of novelty and excitement; they are: the semiotics of the image and horror literature. The research problem can be defined by the question that searches for the nature of the encounter between method and procedure. In other words, how suitable is the semiotic approach for analyzing Arab novelist discourse and revealing its secrets? The conclusions lie in completing the accumulated knowledge to enhance the energies of the different genres and in an attempt to cross-fertilize some genres capable of producing their fruits due to valid hybridization in a place where procedure and employment interact.
У статті досліджуються лінгвістичні особливості англомовних шлюбних оголошень у контексті соціолінгвістичних та лінгвокультурних факторів, які впливають на їхнє конструювання, а згодом і дистрибуцію на інтернет-сайтах та додатках для знайомств. Для аналізу було обрано такі популярні додатки для знайомств як Tinder та Badoo.
Різноманітність у змісті шлюбних оголошень, поглядах кандидатів на шлюб, їхніх очікуваннях від іншої особи та способах представити себе відображає швидкі та драматичні трансформації, які відбуваються в соціальному та культурному житті суспільства.
У роботі досліджуються лексичний склад, стилістичні особливості та комунікативні стратегії, які автори шлюбних оголошень використовують для репрезентації себе та своїх бажань, а також вираження своїх очікувань від потенційного партнера. Віртуальні платформи для знайомств, такі як сайти та додатки, надають користувачам можливість свідомо формувати свій образ та керувати ним. Це особливо важливо у контексті проаналізованих шлюбних оголошень, де мова стає інструментом для створення бажаного враження, а комунікація визначається як стратегічна.
Важливість ефективної комунікації полягає в забезпеченні належного сприйняття повідомлення та точного розуміння інформації, яка у ньому передається. Комунікативна взаємодія у контексті шлюбних оголошень сприяє побудові бажаних для авторів шлюбних оголошень стосунків.
Автори англомовних шлюбних оголошень прагнуть досягти мети комунікації, яка полягає у пошуку людини, яка може стати як романтичним партнером, так і виконувати інші ролі у житті субʼєкта оголошення. У статті досліджуються соціолінгвістичні та лінгвокультурні особливості цього явища, з фокусом на те, як мовні прийоми сприяють управлінню враженнями у цифровому просторі.
Discourse analysis, Computational linguistics. Natural language processing
فرناز ساسانی, مرجان فرجاه, سپیده نواب زاده شفیعی
et al.
در پژوهش حاضر در صدد آن هستیم تا با در نظر گرفتن تفاوتهای زبانی و فرهنگی، به چگونگی انتقال پندارههای فرهنگى در ترجمۀ ادبی بپردازیم. از این رو بهعنوان پیکرۀ این پژوهش، نمونههایی از ترجمههای دو اثر از اریک امانوئل اشمیت را در دورههای مختلف مورد مطالعه قرار خواهیم داد و روشهایی را که مترجمان برای ترجمۀ عناصر فرهنگی به کار بردهاند، با تکیه بر نظریه پیتر نیومارک بررسی خواهیم کرد تا کاربردپذیری این نظریه در محوریت زمان و دورنمای کارآمدی آن براساس سه شاخصۀ اهداف، امکانات و موانع تبیین گردد. با گسترش تسلط بر هوش مصنوعی و بهکارگیری نسل جدیدی از نرمافزارهای ترجمه توسط مترجمان جوان که همواره نسخههای دقیقتری از محتوای اصلی را به زبانهای بیشتری ارائه میدهند، شاهد ظهور نسلی از مترجمان هستیم که روزآمدی نظریههای ترجمه را در خصوص بازنمایی بنمایههای فرهنگی نادیده میانگارند. از آنجا که معنای واژهها یا عبارات نشأتگرفته از فرهنگ یک زبان است، درک معنا مستلزم شناخت کاملی از فرهنگ زبان مبدأ است و به نظر میرسد تا به امروز نرمافزارهای ترجمه بهتنهایی نمیتوانند برای دست یافتن به ترجمۀ دقیق مؤثر واقع شوند. لذا در این جستار سعی بر آن داریم تا اهمیت کاربرد نظریههای ترجمه را در حیطۀ ترجمۀ ادبی بهویژه در ارتباط با ترجمۀ عناصر فرهنگی برای نسل آینده از مترجمان مطرح سازیم.
Language. Linguistic theory. Comparative grammar, Indo-Iranian languages and literature
Large language models (LLMs) have achieved success in acting as agents, which interact with environments through tools such as search engines. However, LLMs are optimized for language generation instead of tool use during training or alignment, limiting their effectiveness as agents. To resolve this problem, previous work has first collected interaction trajectories between LLMs and environments, using only trajectories that successfully finished the task to fine-tune smaller models, making fine-tuning data scarce and acquiring it both difficult and costly. Discarding failed trajectories also leads to significant wastage of data and resources and limits the possible optimization paths during fine-tuning. In this paper, we argue that unsuccessful trajectories offer valuable insights, and LLMs can learn from these trajectories through appropriate quality control and fine-tuning strategies. By simply adding a prefix or suffix that tells the model whether to generate a successful trajectory during training, we improve model performance by a large margin on mathematical reasoning, multi-hop question answering, and strategic question answering tasks. We further analyze the inference results and find that our method provides a better trade-off between valuable information and errors in unsuccessful trajectories. To our knowledge, we are the first to demonstrate the value of negative trajectories and their application in agent-tunning scenarios. Our findings offer guidance for developing better agent-tuning methods and low-resource data usage techniques.
AbstractSpanish has two forms to introduce comparative standards: que ‘that’ and de ‘of.’ The comparative morpheme is always the same más ‘-er/more.’ While que-comparatives show no variation in their syntactic properties, there is significant variation within de-comparatives regarding extraposition, scope, ACD resolution and the syntax of comparative numerals. Despite this variation, I argue that a uniform account is possible. I propose that más has the same syntax across the board (i.e. it takes the late-merged standard as complement, Bhatt and Pancheva 2004) and semantically it is a generalized quantifier over degrees (Heim 2001). The analysis (i) ensures that más and the standard form a constituent, (ii) allows for inverse scope, ACD resolution inside the standard of comparison and extraposition.
Chinese Grammatical Error Correction (CGEC) has been attracting growing attention from researchers recently. In spite of the fact that multiple CGEC datasets have been developed to support the research, these datasets lack the ability to provide a deep linguistic topology of grammar errors, which is critical for interpreting and diagnosing CGEC approaches. To address this limitation, we introduce FlaCGEC, which is a new CGEC dataset featured with fine-grained linguistic annotation. Specifically, we collect raw corpus from the linguistic schema defined by Chinese language experts, conduct edits on sentences via rules, and refine generated samples manually, which results in 10k sentences with 78 instantiated grammar points and 3 types of edits. We evaluate various cutting-edge CGEC methods on the proposed FlaCGEC dataset and their unremarkable results indicate that this dataset is challenging in covering a large range of grammatical errors. In addition, we also treat FlaCGEC as a diagnostic dataset for testing generalization skills and conduct a thorough evaluation of existing CGEC models.
We propose a linguistic interpretation of three-way decisions, where the regions of acceptance, rejection, and non-commitment are constructed by using the so-called evaluative linguistic expressions, which are expressions of natural language such as small, medium, very short, quite roughly strong, extremely good, etc. Our results highlight new connections between two different research areas: three-way decisions and the theory of evaluative linguistic expressions.
COOL is an Object-Oriented programming language used to teach compiler design in many undergraduate and graduate courses. Because most students are unfamiliar with the language and code editors and IDEs often lack the support for COOL, writing code and test programs in COOL are a burden to students, causing them to not fully understand many important and advanced features of the language and compiler. In this tool paper, we describe COOLIO,an extension to support COOL in the popular VSCode IDE. COOLIOprovides (i) syntax highlighting supports for the COOL language through lexing and parsing, (ii) semantics-aware autocompletion features that help students write less code and reduce the burden of having to remember unfamiliar COOL grammar and syntax, and (iii) relevant feedback from the underlying COOL interpreter/compiler (e.g., error messages, typing information) to the students through VSCode editor to aid debugging. We believe that COOLIO will help students enjoy writing COOL programs and consequently learn and appreciate more advanced compiler concepts.
Language models are typically evaluated on their success at predicting the distribution of specific words in specific contexts. Yet linguistic knowledge also encodes relationships between contexts, allowing inferences between word distributions. We investigate the degree to which pre-trained Transformer-based large language models (LLMs) represent such relationships, focusing on the domain of argument structure. We find that LLMs perform well in generalizing the distribution of a novel noun argument between related contexts that were seen during pre-training (e.g., the active object and passive subject of the verb spray), succeeding by making use of the semantically-organized structure of the embedding space for word embeddings. However, LLMs fail at generalizations between related contexts that have not been observed during pre-training, but which instantiate more abstract, but well-attested structural generalizations (e.g., between the active object and passive subject of an arbitrary verb). Instead, in this case, LLMs show a bias to generalize based on linear order. This finding points to a limitation with current models and points to a reason for which their training is data-intensive.s reported here are available at https://github.com/clay-lab/structural-alternations.
The translation of references specific to a given source culture has long been a prominent, and often problematic aspect of translation practice and research. In indirect translation, or the translation of already translated material, linguistic and cultural differences accumulate, meaning that the omission of cultural references (CRs) or culture-specific items (CSIs) might be a generally expected outcome. Yet before such hypotheses can be tested, research methods are needed that can account for broad patterns across whole texts, and preferably, across semantic categories, genres, time periods, and languages. A ‘textual’ approach, focused on the linguistic context in which CRs are likely to occur, should complement the currently dominant ‘cultural’ approach, which mainly relies on predefined categories and intuition for the selection of objects of study. This article illustrates that corpus research, and particularly keyness analysis, can aid in uncovering recurrent structural patterns and textual functions in which CRs are expected to pose translation difficulties. In this regard, it focuses on expressions of enumeration, or lists, and indicators of identification, or voice. Based on a trilingual (English, French, and Italian) corpus-assisted study of Jonathan Swift’s Gulliver’s Travels (1726) and John Cary’s An Essay on the State of England (1695), the article accentuates the productive complementarity of numerical operations and context-sensitive readings.
Language and Literature, Language. Linguistic theory. Comparative grammar
Julieta Aguilera Hernández, María Elena Hidalgo Mora
La pertinencia de las crónicas de viaje como fuente para las
ciencias históricas constituye un asunto pendiente para la historiografía
santiaguera. El presente artículo ofrece un bosquejo panorámico sobre
las crónicas de viaje que reflejaron múltiples aspectos de la vida cotidiana
de la ciudad de Santiago de Cuba entre 1800 y 1868. A partir de
las principales temáticas abordadas por los viajeros —cuyas memorias
tuvieron mayor divulgación en los predios local y nacional—, se valida el
significado de estos textos como fuente para los estudios historiográficos
locales del período colonial.
The relevancy of the trip chronicles as a source for the historical
sciences constitutes a pending matter for the historiography of
Santiago de Cuba. The present article offers a panoramic outline on the
trip chronicles that reflected multiple aspects of the daily life of the city
of Santiago from Cuba between 1800 and 1868. Starting from the main
topics approached by travelers –whose memories had wider dissemination
in the local and national scenery—, it´s has been validated the
meaning of these texts like a source for the local historiographical studies
of the colonial period.
Philology. Linguistics, Language. Linguistic theory. Comparative grammar
Many studies have investigated isolated dimensions of learning styles (e.g. field independence/dependence) for their role in foreign language learning, but relatively few studies have used a comprehensive learning styles instrument to determine predictors of language learning strategies used by students. Hence, employing the descriptive and correlational method, this study aimed to identify students’ minor, major, and negligible learning styles, students’ usage of language learning strategies, the difference in the learning styles and language learning strategies based on gender, and the relationships among those three variables. A total of 30 students enrolling in the first year of senior high school were given two kinds of questionnaire, the Indonesian version of PLSQ and SILL. The result revealed gender differences only occurs in compensation strategy, in favor of female students. Furthermore, the correlational study revealed significant relationships between visual style and cognitive and metacognitive strategies; between auditory style and cognitive and compensation strategies. Moreover, social strategies are correlated with tactile, group, and individual styles. These findings are useful for both teacher and student to employ strategies suitable with their learning styles.
Language. Linguistic theory. Comparative grammar, English language