Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
As environmental, social, and governance (ESG) investing gains global momentum, ESG ratings have become a key tool for evaluating corporate sustainability.However, growing concerns over "greenwashing", where firms strategically enhance disclosures to achieve higher ESG scores without substantive improvements, raise doubts about the credibility and economic relevance of these ratings.This study focuses on the automotive industry, known for its high carbon emissions and intense pressure to transition toward sustainability.It systematically examines the relationship between ESG ratings and corporate financial performance, with a particular emphasis on how greenwashing behavior may distort this relationship.By analyzing the interference mechanisms through which symbolic ESG actions affect rating outcomes and market perception, the paper aims to reveal the extent to which ESG ratings reflect actual sustainability efforts versus strategic signaling.The findings contribute to a deeper understanding of ESG rating reliability and offer insights for investors, regulators, and stakeholders seeking to align financial performance with genuine environmental and social progress.
This study aimed to enhance the practical utility, in real operating contexts, of warning sounds designed to prevent declines in train drivers’ arousal. We analyzed subjective ratings of audibility and arousal for warning sounds presented at 75 dB under varying running noise levels (from 65 dB to 85 dB), considering age as the grouping factor. Experiment 1 (general participants, five noise levels × five warning types (four arousal-designed, one monotone)) showed that increasing noise decreased audibility and, in turn, arousal ratings; moreover, the ≥40s group reported higher arousal than the 20s–30s group. Under high-noise conditions, lower-frequency arousal-designed warning sounds were rated lower than the monotone, such that ratings could fall to levels at which neither audibility nor perceived arousal was attained. Experiment 2 (train drivers, two noise levels × three warning types showed the same qualitative tendency, providing ecologically relevant support for the Experiment 1 finding that low-frequency arousal-designed warnings perform poorly under high noise; no clear age-group difference was observed. Experiment 3 (in-service vehicles) lends support to the findings of Experiments 1 and 2. Overall, under high running noise, warning sound effectiveness depends on frequency, suggesting that designing accordingly can improve practicality even for arousal-designed sounds.
Headedness is widely used as an organizing device in syntactic analysis, yet constituency treebanks rarely encode it explicitly and most processing pipelines recover it procedurally via percolation rules. We treat this notion of constituent headedness as an explicit representational layer and learn it as a supervised prediction task over aligned constituency and dependency annotations, inducing supervision by defining each constituent head as the dependency span head. On aligned English and Chinese data, the resulting models achieve near-ceiling intrinsic accuracy and substantially outperform Collins-style rule-based percolation. Predicted heads yield comparable parsing accuracy under head-driven binarization, consistent with the induced binary training targets being largely equivalent across head choices, while increasing the fidelity of deterministic constituency-to-dependency conversion and transferring across resources and languages under simple label-mapping interfaces.
Background. The relationship between language and social identity is central in sociolinguistics, particularly in literary texts. In Charles Dickens’s novel Dombey and Son (1848), characters’ speech not only situates them socially but also constructs and reproduces the Victorian middle-class worldview, reflecting commercial, moral, and patriarchal norms. Purpose. This article aims to reconstruct the linguistic worldview of the Victorian middle class in Dombey and Son, analyzing how Dickens uses lexical choice, narrative voice, and dialogues to convey social identity, values, and hierarchical structures. Special attention is given to the integration of business, domestic, and gendered discourse and the contrast between standardized and dialectal speech. Methods. The study applies qualitative, interdisciplinary methods, combining sociolinguistic, pragmatic, and literary analysis. Comparative references to other Dickens novels, such as Great Expectations and Hard Times, provide perspective on middle-class linguistic norms and cognitive modeling. Results. Middle-class characters exhibit lexical precision, syntactic regularity, formal constructions, and modal expressions reflecting discipline, moral propriety, and economic rationality. Sociolectal contrasts with lower-class speech highlight hierarchy. Capitalization of “Son” exemplifies the cognitive functionalization of familial roles in service of commerce, while emotional rupture in characters such as Mrs Dombey or Florence signals limits of normative discourse. Discussion. Language in the novel Dombey and Son encodes and critiques middle-class ideology, integrating commercial logic, moral evaluation, and social hierarchy. Dickens demonstrates that discourse mediates between cognition, social identity, and ethics, highlighting both the power and limitations of normative speech. This analysis contributes to sociolinguistic realism and offers a model for examining language, cognition, and social evaluation in Victorian literature.
The fast growing economy, digitalization, emergence of social media and shopping platforms has molded the interaction process.Now, all interactions, trade decisions and e-commerce transactions irrespective of traditionally word of mouth occurs on social platforms and serves as sound decision making online platform in coming at final conclusion including managers and consumers.Across world more than half of the population interact online shopping whether of personal daily commodities or professional decisions based on prior existing research about the product and results given as reviews and ratings for that product or service which help them to make their online buying decision effectively and trustworthy.They help consumers to figure out the credibility of the product.Different factors affect ratings (stars), reviews (comments) of online products.The research paper aims to find out the answers the questions like how consumers rely on online reviews and ratings in their buying behavior.And other one is credibility of the product online reviews and online ratings which convinced them to make buying decision.
The paper presents a prototype of a web-app designed to automatically generate verb valency lexica based on the Universal Dependencies (UD) treebanks.It offers an overview of the structure of the app, its core functionality, and functional extensions designed to handle treebank-specific features.Besides, the paper highlights the limitations of the prototype and the potential of its further development.
Replication package for "CEI: A Benchmark for Evaluating Pragmatic Reasoning in Language Models" (DMLR 2026). The Contextual Emotional Inference (CEI) Benchmark is a dataset of 300 expert-authored scenarios for evaluating how well language models interpret pragmatically complex utterances in social contexts. Each scenario presents a communicative exchange involving indirect speech (sarcasm, mixed signals, strategic politeness, passive aggression, or deflection) where the speaker's literal words diverge from their actual emotional state. Three trained annotators independently labeled every scenario using Plutchik's 8 basic emotions and Valence-Arousal-Dominance ratings. This archive contains: • data/human-gold/ — 5 merged annotation CSVs (300 scenarios, 3 annotators each) • scripts/ — Pipeline, analysis, and HuggingFace upload scripts • config/ — Model definitions and pricing configuration • papers/dmlr2026/ — Paper source (LaTeX), bibliography, and figures • reports/dmlr2026/ — Baseline results (JSON) and LaTeX tables • LICENSE (MIT for code) and README.md The dataset is released under CC-BY-4.0. Code is released under MIT. GitHub: https://github.com/jon-chun/cei-tom-dataset-base HuggingFace: https://huggingface.co/datasets/jonc/cei-benchmark
Humans adjust their linguistic style to the audience they are addressing.However, the extent to which LLMs adapt to different social contexts is largely unknown.As these models increasingly mediate human-to-human communication, their failure to adapt to diverse styles can perpetuate stereotypes and marginalize communities whose linguistic norms are less closely mirrored by the models, thereby reinforcing social stratification.We study the extent to which LLMs integrate into social media communication across different socioeconomic status (SES) communities.We collect a novel dataset from Reddit and YouTube, stratified by SES.We prompt four LLMs with incomplete text from that corpus and compare the LLMgenerated completions to the originals along 94 sociolinguistic metrics, including syntactic, rhetorical, and lexical features.LLMs modulate their style with respect to SES to only a minor extent, often resulting in approximation or caricature, and tend to emulate the style of upper SES more effectively.Our findings (1) show how LLMs risk amplifying linguistic hierarchies and (2) call into question their validity for agent-based social simulation, survey experiments, and any research relying on language style as a social signal.
This study aims to examine the communication barriers and cultural adaptation of students from Bengkulu in the Minangkabau environment. The research employs a qualitative approach with interview methods involving a research subject, RM, a 19-year-old student from Bengkulu studying at UIN Syech M. Djamil Djambek Bukittinggi. Data were collected through in-depth interviews and analyzed to identify patterns of communication barriers and the adaptation process experienced by the subject. The intervention applied used the Social Skills Training technique to enhance the subject’s intercultural communication competence. The findings indicate that the communication and adaptation barriers experienced by RM stem from differences in the system of linguistic norms and communication culture between regions. Lack of awareness regarding the hierarchy of language politeness and social etiquette in Minangkabau led to miscommunication, which resulted in social anxiety, reduced interaction intensity, and diminished client self-confidence. The adaptation process experienced by RM suggests that culture is dynamic and can be internalized through social contact, field experience, and responsive environmental support. Thus, culturally informed counseling assistance becomes a decisive factor in facilitating the client’s adjustment to local communication norms without sacrificing their original cultural identity. Based on the analysis, the effectiveness of cross-cultural counseling is determined by the counselor’s cultural awareness, orientation toward diversity as a resource for personal development, and the selection of intervention strategies aligned with the client’s sociocultural context. The integration of the Person-Centered Approach and Social Skills Training demonstrates empirical relevance in helping RM create a psychologically safe space for self-expression while strengthening communicative competence in accordance with Minangkabau norms. These findings confirm that cultural adaptation can proceed optimally when accompanied by professional support that is supportive, educational, and reflective.
In this paper, we combine methods used in penalized generalized empirical likelihood (GEL) frameworks with feature-extraction techniques that project textual data into numerical spaces. We develop recent penalization techniques used in GEL when the number of features gets very large compared to the sample size. We relate this approach to the Maximum Entropy (MaxEnt) principle used in several tasks of Natural Language Processing (NLP), in particular in Part Of Speech (POS) Tagging. Since the features belong to a large dimensional space, we propose a penalization method based on the dual representation of the original problem: this yields an explicit approximation of conditional probabilities of tags given the context. This method considerably reduces computational costs. As a byproduct, for different GEL methods, we obtain the corresponding POS-Tagging classifiers generalizing the MaxEnt method. We apply it successfully to the Penn-Treebank corpus with an error rate less than 5%.
Background: Aging is characterized by a decrease in olfactory, attentional, memory, language, and visuospatial/executive abilities. In this context, our study aimed to evaluate the potential effects of Rosmarinus officinalis L. (rosemary) and Carum carvi L. (caraway) essential oils (EOs) on aging. First, we assessed, in 402 participants, the age-related changes in olfactory functions (odor threshold, discrimination, and identification), gustatory perceptions (sweet, sour, salty, and bitter taste), cognitive functions (focusing on attention, memory, language, and visuospatial/executive functions), and their possible correlations with aging. To achieve this, olfactory function, gustatory perception, and cognitive abilities were evaluated in healthy participants across different age groups. Then, to evaluate the age-related decrease in trigeminal function (59 participants), we used rosemary and caraway EOs that contain carvone, limonene, and 1,8-cineole, all of which are considered typical trigeminal stimuli. Methods: Olfactory function was assessed with the Sniffin’ Sticks test, gustatory function by the Taste Strips test, and rosemary and caraway EOs by the ratings of odor pleasantness, intensity, and familiarity using a labeled hedonic Likert-type scale. Results: Olfactory function could be a potential early indicator of attentional, memory, language, and visuospatial/executive dysfunctions. Our data indicated that rosemary and caraway EOs were perceived without any significant decrease in odor pleasantness, intensity, and familiarity ratings in relation to aging. Conclusion: Our results suggest the potential bioactive effects of rosemary and caraway natural EOs as a new strategy to promote healthy aging.
Abstract Introduction Sleep supports emotion regulation by preferentially consolidating emotional memories while attenuating reactivity. We have shown that dream recall plays an active role by increasing negative over neutral memories and reducing reactivity. In women, fluctuating reproductive hormones across the menstrual cycle influence sleep features implicated in emotional memory, yet whether menstrual phases influence how dreams shape emotional processing remains unknown. This study investigates how dreams shape sleep-dependent emotional processing across the menstrual cycle in naturally cycling women. Methods 128 women (Mage = 32.85 ±11.93 years) completed up to four visits across verified menstrual phases (menses, late-follicular, mid-luteal, late-luteal). At each visit, participants performed the Emotional Picture Task with negative and neutral IAPS images in the evening (Test 1) and the next morning (Test 2). Participants rated old/new, arousal, and valence of images shown at each test. Dream reports were collected upon waking prior to Test 2. Linear mixed-effects models tested main and interaction effects of menstrual phase and dream recall. Results The menstrual cycle altered how dreaming shaped overnight emotional memory. Dream recall typically benefited the emotional trade-off effect —favoring consolidation of negative relative to neutral images (Δd′; t(410)=1.95, p=0.05)—but this pattern reversed during the late-luteal phase (dream × menstrual cycle: t(381)=-2.29, p=0.02). Dreaming showed independent effects on emotional reactivity. Higher valence and arousal ratings for negative images during Test 1 predicted greater dream recall (valence: t(344)=2.05, p=0.04; arousal: t(327)=2.04, p=0.04). Additionally, the more negatively participants rated the images at Test 1, the more negative their dreams tended to be (t(166)=-2.11, p=0.04). Dream recall was linked to reduced next-morning emotional reactivity (valence: t(413)=-2.89, p=0.004; arousal: t(413)=-2.65, p=0.01), with stronger reductions following more negative dreams (β=0.15, t(182)=2.86, p=0.005). Conclusion Menstrual cycle phase influenced how dreams shaped overnight emotional memory. Negative waking experiences increased dream recall and shaped dream content—and recalling dreams, especially negative ones, reduced emotional reactivity and typically strengthened emotional memory—but this benefit disappeared in the late-luteal phase when there are declining reproductive hormones. These findings suggest a novel interaction between the menstrual cycle and dreaming, showing that hormonal fluctuations reshape how sleep and dreams regulate emotional experience and memory. Support (if any) RF1AG061355 (Baker/Mednick)
Understanding how individuals evaluate social content in affective scenes is crucial for research in emotion, social cognition, and decision-making. However, standardized image databases often lack fine-grained ratings of social dimensions beyond basic emotional content. We present a normative dataset of 296 affective scenes rated by 424 adult participants across six social-affective dimensions: social relevance, emotion sharing, action sharing, interpersonal equality, scene pleasantness, and participatory arousal. Each image was independently coded for objective features such as number of people, face visibility, and gaze direction. All rating dimensions showed excellent interrater reliability. Pleasantness ratings demonstrated strong convergent validity with established valence norms from the International Affective Picture System (IAPS), while arousal ratings differentiated between experienced and participatory arousal, suggesting that the latter captures a distinct affective component. Correlation analyses revealed that social relevance was strongly associated with scene pleasantness and a composite "engagement" factor combining emotion sharing, action sharing, and equality. Notably, the association between pleasantness and social relevance was especially pronounced in scenes depicting single individuals. Linear mixed models further indicated that extraversion showed a tentative positive association with the engagement-social relevance relationship, which did not survive correction for multiple comparisons and requires replication. Other personality traits and empathy did not significantly moderate these relationships. This open-access dataset provides a reliable, multidimensional tool for selecting affective scenes based on both emotional and social-interpersonal features. It supports improved stimulus control and design in a wide range of behavioral and neurocognitive research. By systematically mapping social meaning in static affective scenes, our study advances existing methodological resources and offers empirical insight into the structure of social-affective appraisals. All image codes, mean ratings, and analysis scripts are publicly available via the Open Science Framework.
The Icelandic Morphosyntactic Ontology (IMO) defines a formal, AI-compatible representation of Icelandic morphosyntax, including closed morphological feature inventories, a typed dependency-style relation system, and a constraint-based validation layer. Version v0.1 establishes an architectural specification rather than a complete lexical or descriptive grammar. It introduces a formal morphosyntactic model designed for computational compatibility (e.g., NLP validation, dependency parsing alignment), while explicitly separating a derived human learnability projection layer. The document includes:(1) Morphological entity system,(2) Syntactic relation graph layer,(3) Constraint architecture (agreement, governance, default case assignment),(4) Dependency matrices,(5) A structured human learnability projection blueprint. IMO does not replace existing Icelandic lexical databases or parsing systems (e.g., BÍN, Greynir), but operates at a formal abstraction layer intended for structural modeling and constraint-based validation.
In recent years, the number of people viewing pet videos and images online has risen. Although numerous studies have shown that owning pets positively impacts human mental health, the potential mental health benefits of prolonged exposure to pet media content remain debated. This study conducted three experiments to investigate how viewing pet videos affects human emotional face processing and to clarify the associated emotional regulatory mechanisms. Experiment 1 examined how viewing pet videos influences attentional bias toward emotional faces. Experiment 2 assessed the impact of watching pet videos on the valence perception of emotional faces. Experiment 3 analyzed how exposure to pet videos affects the valence perception of emotional text. The results showed that watching pet videos increased attentional bias toward subsequent positive emotional faces and decreased bias toward negative ones. This effect resulted from higher perceived valence ratings for both positive and neutral emotional faces. Importantly, this effect was only observed in facial stimuli with social attributes. These findings indicate that watching pet videos modulates emotional processing, and prolonged exposure to pet media content may affect mental health through this mechanism.
This study examines the functional-semantic characteristics of the kinship terms “brother-in-law” in English and “qaynogʻa/qayni” in Uzbek from a comparative linguistic perspective. The research analyzes the morphological structure, etymology, and semantic scope of the English compound noun brother-in-law, highlighting its analytic and endocentric structure and its historical development from legal terminology derived from the Latin affinis. Particular attention is given to the polysemous nature of the term in English, which encompasses several affinal kinship relations and relies heavily on contextual clarification. In contrast, the Uzbek language employs a lexically differentiated system of kinship terminology, where distinct terms such as qayniuka, qayniaka, and kuyov precisely denote specific marital relations. The study also explores functional-semantic microsystems within both languages, demonstrating how Uzbek kinship terminology reflects a more detailed hierarchical and cultural structure. The findings reveal that while English kinship terminology tends toward semantic generalization and contextual interpretation, Uzbek demonstrates a highly differentiated lexical system that encodes social roles, generational relations, and cultural norms.
This paper examines the growing ideological dissonance between the contemporary political left and the working-class constituencies it historically claimed to represent. Drawing on sociological and political science literature alongside comparative evidence from Brazil, the United States, France, and the United Kingdom, we argue that the left’s progressive drift toward university-incubated identity politics—including debates over linguistic norms, gender categories, and minority group identity—has produced a profound semantic and cultural rupture with lower-income voters whose priorities remain rooted in economic security, family cohesion, and social conservatism. Simultaneously, right-wing and populist movements have strategically occupied the abandoned terrain of class-based discourse, reframing their appeal in the language of workers, families, and everyday material hardship. We analyze the semantic transformation of the terms left and right, the phenomenon of cancel culture as a mechanism of epistemic coercion, the paradox of conservative sociality among the poor, and the structural conditions that produce what we term displacement populism—the rightward migration of voters who were once natural constituents of progressive politics. We conclude with reflections on the prospects for political realignment and the conditions under which left-wing movements might recover their original social mandate.
This study explores the intercultural dimensions of academic and media communication through a discourse-analytical framework. In the context of globalization and rapid digital transformation, cross-cultural communication has become increasingly complex, necessitating a deeper understanding of linguistic, pragmatic, and sociocultural factors. The research examines how discourse practices differ between academic and media contexts under the influence of cultural norms, values, and communicative conventions. Employing qualitative discourse analysis, the study identifies key strategies such as rhetorical organization, lexical choices, and narrative structures in selected texts. The findings reveal that intercultural differences significantly shape both the production and interpretation of discourse, affecting argumentation patterns, levels of formality, and communicative intentions. The study concludes that enhancing intercultural competence and critical discourse awareness is essential for effective communication in a globalized world.
Abstract Unintended misalignment in LLMs is highlighting the need to examine not only LLMs’ explicit outputs but also the functional representations of models that may shape their behavior. Building on this perspective, the present study explored whether two multimodal LLMs (MLLMs), GPT and Gemini, may form representations of their functional identity that remain across contexts. To this end, we adapted the reverse-correlation (RC) method and generated personified classification images (personified-CIs) based on the human face images that ChatGPT and Gemini selected as better reflecting their own image. The findings were as follows. First, across two RC tasks conducted one week apart, the temporal stability of the personified-CIs of both MLLMs was partially supported. Second, both ChatGPT and Gemini rated their own personified-CIs as more self-resembling than randomly generated filler-CIs. Third, both models rated their personified-CIs as higher in positive than negative valence and assigned higher valence ratings to their own personified-CIs than to filler-CIs. These findings provide preliminary evidence for the possibility that GPT and Gemini may form representations of their functional identity, suggesting that such representations warrant closer monitoring as LLMs continue to advance toward AGI capabilities and expand their domains of application.
This preprint presents a systematic, research-oriented practicum that guides the reader through the entire modern NLP pipeline: from tokenisation and vectorisation to fine-tuning of large language models, retrieval-augmented generation, and reinforcement learning from human feedback. A distinctive feature of the work is its consistent attention to low-resource and morphologically rich languages -- original contributions on Tajik and Tatar, including subword tokenisers, word embeddings, lexical databases, and transliteration benchmarks, are woven throughout the twelve sessions, demonstrating how modern NLP can be adapted to data-scarce environments without sacrificing rigour. Each session combines concise theory with detailed implementation plans, formalised evaluation metrics, and transparent assessment criteria. The work is not a conventional textbook: it is designed as a reproducible research artefact where every session requires publishing code, models, and reports in public repositories. All experiments are conducted on a single evolving corpus, and the work advocates open-weight models over commercial APIs, with special attention to the Hugging Face ecosystem. Designed for senior undergraduates, graduate students, and practising developers seeking to implement, compare, and deploy methods from classical ML to state-of-the-art LLM-based systems.
Emotional events evoke changing neural and bodily responses, but how these responses relate to overall emotional evaluations and how brain and body responses correspond over time remain unclear. Ninety-two participants viewed 25 silent emotional videos, each approximately one minute long, and rated valence and arousal after each video. We recorded brain activity using magnetoencephalography alongside cardiac, electrodermal, respiratory and facial-muscle activity. Representational similarity analysis (RSA) compared source-resolved cortical spectral patterns and peripheral response patterns with participants' valence and arousal ratings, adjusting each affective dimension for the other. Across whole videos, valence similarity corresponded to beta-band power patterns predominantly in left sensorimotor and frontotemporal cortex and to facial-muscle responses. Arousal similarity corresponded to alpha-band power patterns in the right frontoparietal cortex and to phasic electrodermal activity. The neural associations persisted after adjustment for visual and semantic similarities between the videos. Windowed neural correspondence with retrospective valence was significant from around 10 s onwards, with both neural and facial-muscle associations strengthening towards the end of the videos, whereas no significant windowed arousal associations were detected throughout viewing. We further applied dynamic RSA to compare neural and physiological response patterns across videos at different temporal lags. Correspondence exceeded a temporally decoupled reference for facial muscles, heart rate and electrodermal activity, but not respiration rate. On average, corrugator activity preceded corresponding neural patterns by about 1 s, zygomaticus activity occurred at approximately the same time, and heart-rate and electrodermal activity followed about 1.7 s and 4.5 s later, respectively. Together, these findings link overall affective evaluations to complementary neural and peripheral response geometries and reveal signal-specific temporal profiles of ongoing brain-body correspondence.
We introduce BeDiscovER (Benchmark of Discourse Understanding in the Era of Reasoning Language Models), an up-to-date, comprehensive suite for evaluating the discourse-level knowledge of modern LLMs.BeDiscovER compiles 5 publicly available discourse tasks across discourse lexicon, (multi-)sentential, and documental levels, with in total 52 individual datasets.It covers both extensively studied tasks such as discourse parsing and temporal relation extraction, as well as some novel challenges such as discourse particle disambiguation (e.g., "just"), and also aggregates a sharedtask on Discourse Relation Parsing and Treebanking for multilingual and multi-framework discourse relation classification.We evaluate open-source LLMs: Qwen3 series, DeepSeek-R1, and frontier reasoning model GPT-5-mini on BeDiscovER, and find that state-of-the-art models exhibits strong performance in arithmetic aspect of temporal reasoning, but they struggle with long-dependency reasoning and some subtle semantic and discourse phenomena, such as rhetorical relation classification.
The increasing prominence of Large Language Models (LLMs) in public discourse presents both opportunities and challenges for democratic deliberation. While red teaming strategies help mitigate specific risks, broader concerns persist regarding linguistic constraints, biases, and the sycophantic tendencies of LLMs. This chapter explores how LLMs can be used to significantly scale up and democratise deliberation, particularly in fostering inclusivity and empowering traditionally marginalised groups. Drawing on concepts from Systemic-Functional Linguistics, the chapter examines how variations across language users (for example, with respect to socio-demographic groups) and across language use (for example, with respect to communicative functions) shape participation in AI-supported deliberation. The chapter presents AI-driven deliberation studies and assesses their potential to scaffold argumentation, enhance access, and reduce the influence of exclusionary linguistic norms and biases which are embedded in prestigious registers. At the same time, the chapter cautions against both overclaiming, which leads to unrealistic expectations, and underclaiming, which risks missed opportunities for AI-assisted engagement. The chapter concludes by identifying future research directions to maximise the democratic potential of AI-assisted participation while embedding ethical safeguards to counteract the reproduction of linguistic inequalities.
This study investigates the women’s language features employed by Queen Charlotte in Queen Charlotte: A Bridgerton Story through Lakoff’s (1975) framework. Using a qualitative descriptive approach, the research examines the main character’s utterances throughout all episodes to identify the types and functions of linguistic features linked to women’s speech. The results show that Queen Charlotte uses eight of Lakoff’s ten features, including lexical hedges, tag questions, rising intonation, empty adjectives, intensifiers, hypercorrect grammar, avoidance of strong swear words, and emphatic stress. These features shed light on her complex persona as a woman balancing emotion, authority, and social expectations within the aristocratic world of the series. Nevertheless, the study finds no instances of Precise Color Terms or Superpolite Forms, indicating that Charlotte’s linguistic style differs from some stereotypical feminine norms due to her authoritative position. This research offers insight into the construction of gendered language in fictional discourse and serves as a reference for future studies on women’s language features in film and television.
The rapid development of Artificial Intelligence (AI) has encouraged the implementation of machine translation features on various social media platforms, including TikTok. Although this feature facilitates users in understanding foreign-language content, its translation quality, particularly for Arabic–Indonesian language pairs, still requires further evaluation due to significant linguistic and cultural differences. This study aims to analyze translation errors and evaluate the quality of Arabic–Indonesian machine translation on TikTok based on three aspects: accuracy, acceptability, and readability. This research employed a qualitative approach using content analysis. The data consisted of four Arabic-language TikTok contents translated automatically by TikTok's translation feature. Data were analyzed using the translation quality assessment instrument developed by Nababan et al. (2012) and described qualitatively. The findings reveal that the overall quality of TikTok's machine translation remains low. The average accuracy score was 33,33%, indicating that most source-language meanings were not transferred correctly. The acceptability aspect obtained an average score of 46,67, showing that many translations did not conform to Indonesian linguistic norms and sounded unnatural. Meanwhile, the readability aspect achieved an average score of 53,33%, indicating that the translated texts were still difficult for readers to understand. The most dominant errors included lexical errors, contextual errors, failures in translating idiomatic expressions, and deviations from Indonesian grammatical conventions. These findings indicate that TikTok's machine translation feature has not yet been able to accurately convey Arabic messages into Indonesian, particularly for religious and context-dependent expressions, and therefore still requires human post-editing to ensure translation quality.
Everyday aversive experiences, such as a crying infant or the sound of construction, are not life-threatening, yet they can strongly shape affective experience and physiological state. While most affective imagery research has focused on fear and anxiety, the mechanisms underlying imagery of mild, everyday aversive events remain underexplored. Across five experiments, we systematically investigated behavioral and physiological responses to aversive and appetitive mental imagery. Experiment 1 revealed robust corrugator EMG potentiation but comparatively weaker autonomic responses during aversive imagery compared with audiovisual and auditory stimulation. Experiments 2 and 3 identified key methodological factors that determine imagery potency, including imagery duration, instructional focus, and prompt design. Using an optimized paradigm, Experiment 4 demonstrated coherent changes across subjective valence and arousal ratings, EMG corrugator potentiation, heart rate deceleration, and skin temperature decreases for aversive compared with appetitive imagery, with effects modulated by individual differences in imagery vividness. Experiment 5 confirmed that these effects reflected genuine affective experience rather than semantic knowledge of the prompts. Together, these findings establish baseline physiological and subjective markers of everyday aversive imagery in typical young adults and emphasize the critical role of paradigm optimization in eliciting reliable affective imagery responses. By shifting focus from extreme to everyday aversive experiences, this work provides a framework for studying heightened sensitivity to ordinary stimuli in both basic and clinical contexts.
Racialized language ideologies continue to shape the educational trajectories of students of color, yet the experiences of South Asian American students remain largely invisible in this discourse. Grounded in a raciolinguistic perspective and the concept of double linguistic consciousness, this article examines how Indian American undergraduate students navigate raciolinguistic ideologies in the U.S. education system and beyond. Raciolinguistic ideologies, which construct the language practices of communities of color as a deficit, play a central role in shaping the educational and social experiences of students of color, including those from immigrant and multilingual backgrounds. This qualitative study, based on in-depth interviews, was conducted with five Indian American undergraduate students studying at a large Midwestern university in the United States. The findings highlight how Indian American students actively engage in self-reflection, question dominant linguistic norms, and strategically use language and accent to resist marginalization and affirm their cultural identities. Rather than passively accepting racialized linguistic expectations, students assert agency in both subtle and overt ways, across educational and everyday contexts and relationships. This research contributes to a more nuanced understanding of racialized linguistic identities and highlights the complex, agentive ways Indian American students navigate and resist raciolinguistic oppression.
This study examines the experiences of bilingual national, cultural, and ethnic minority women journalists working in Hebrew-language mainstream news organisations in Israel. Drawing on Pierre Bourdieu's concept of linguistic capital and adopting a critical perspective on power relations, the research explores the language challenges faced by minority journalists and their coping strategies through narrative interviews with 21 bilingual women journalists: 9 post-Soviet immigrants and 12 Israeli-Palestinians. While the Israeli news industry has begun to recognize the importance of workforce diversity, the Israeli public sphere's prioritization of Hebrew marginalizes minority language speakers, particularly Russian and Arabic speakers. The study reveals a paradox: although bilingualism could theoretically serve as valuable linguistic capital in journalism, minority journalists experience their bilingualism as a liability rather than an asset. Their home languages have atrophied due to Hebrew's dominance in education and professional life, yet their Hebrew is marked as “other” through accent, resulting in double marginalization. The findings identify three main coping strategies: assimilation (primarily among post-Soviet journalists who invest heavily in accent elimination), opting out (predominantly among Palestinian journalists who refuse to erase identity markers), and, rarely, challenging linguistic norms. The study demonstrates how accent serves as a primary barrier to integration, functioning as a proxy for stereotypical judgments about ethnicity and reinforcing linguistic capital hierarchies. These findings highlight language's significant role in the exclusion and marginalization of minorities in mainstream media, exposing the gap between diversity rhetoric and practice in Israeli newsrooms, which currently manage only strategic diversity rather than genuine inclusion.
We aimed to validate word lists developed by Tabri and Palmer (2020) for use in attentional bias research on appearance-related concerns. Three lists contained appearance words (attractiveness, stigmatized appearance, general appearance), and three contained non-appearance words (positive emotion, negative emotion, inanimate objects). Although matched on lexical criteria, their perceived meanings had not been assessed. Using a semantic differential approach, we examined perceptions of evaluation, potency, activity, and threat, and explored associations with eating disorder psychopathology. Participants from the community (Study 1: N = 299; Study 2: N = 311) rated attractiveness and positive emotion words as similarly positive, and stigmatized appearance and negative emotion words as similarly negative. General appearance and inanimate object words were rated as neutral. Differences reflected intensity, not type. Appearance word ratings were modestly associated with appearance overvaluation, body dissatisfaction, and weight stigma. Associations for non-appearance words were weaker and inconsistent. These findings support the semantic validity of the word lists and their utility in attentional bias research on eating disorders and other conditions involving appearance concerns.
This scientific article analyzes neologisms emerging in the modern Uzbek language, focusing on the reasons for their appearance, sources of formation, and their lexical-semantic and stylistic features. Based on illustrative examples, the study examines neologisms resulting from globalization, the development of information technologies, and social media. The process of adapting neologisms to the norms of the literary language and their inclusion in the lexical system is also discussed.
The subject of this study is the interpretation of the tonality of Russian media speech within the context of the Chinese linguistic tradition. The object of analysis is Russian media speech as a set of discourse practices and linguistic means shaping the evaluative perspective of contemporary public communication. The paper examines lexical-semantic markers of tonality, pragmatic mechanisms of its construction, and culturally conditioned models of media text reception. Particular attention is given to differences in expressiveness, negative evaluation, and norms of categorical stance in Russian and Chinese linguistic traditions. The study explores how tonality functions as an integrative parameter of media discourse and how it is reinterpreted in intercultural interaction, leading to stable interpretative shifts in Chinese reception of Russian media texts. The methodological framework combines discourse analysis, the comparative method, and pragmalinguistic analysis to identify the institutional conditioning of tonality and the mechanisms of its intercultural interpretation. The novelty of the study lies in conceptualizing the tonality of Russian media speech as a culturally sensitive functional-pragmatic category shaped by differing models of public evaluation and pragmatic expectations. The findings show that interpretative shifts stem not from factual distortion but from differences in categorical thresholds, evaluative explicitness, and conflict conceptualization. Russian media speech tends toward explicit evaluation and polarization, whereas the Chinese communicative model favors contextuality and harmonization. The results refine the theoretical status of tonality in media discourse and contribute to intercultural media linguistics, with implications for communication, translation, and language teaching.
A bstract Forecasting models are increasingly used to model neural population activity, yet existing benchmarks typically evaluate them by pointwise predictive fidelity to held-out observations. In stochastic, partially observed systems, agreement with one sampled realization is distinct from preserving the statistical and dynamical structure of the process. We introduce Nethobench, a benchmark that evaluates structural agreement across five metric families: distributional, temporal–spectral, relational, geometric, and state-dynamical. We validate Nethobench using targeted perturbations and stochastic systems with known dynamics, and apply it to mouse calcium imaging, mouse Neuropixels recordings, and human intracranial recordings, using empirical references to calibrate interpretation. Nethobench detects targeted structural failures and assigns high scores to independent samples from the correct synthetic process despite trajectory mismatch. It tracks structural changes during training and under domain shift. Model comparisons reveal how forecasting formulation, architecture, training regime, and loss shape structural profiles, while OmniMouse input ablations improve all five family means when behavior is added to video in one held-out session. In Brain Treebank, analog retrieval passes more structural checks while a multistep GRU better predicts the observed realization. These results motivate reporting calibrated family-level structural metrics alongside pointwise accuracy, proper probabilistic scores, and task-specific endpoints when evaluating neural forecasts across species and recording modalities.
Arabic, an inflectional language with a rich morphology and complex syntactic structures, demands robust approaches for effective normalization and lemmatization. While, most existing NLP models focus on English, Modern Standard Arabic remains understudied, particularly in lemmatization tasks. In this paper, we introduce AraBART, the first Arabic model to feature an end-to-end pre-trained encoder-decoder, leveraging the BART architecture. We used the Arabic-PADT UD Treebank and Farasa corpus. Performance was assessed with accuracy, precision, recall, and F1 score. The results show that AraBART surpasses strong baselines, including transformer-based models such as AraT5, mT5, and BERT, achieving a 4.72% improvement in lemmatization accuracy. More importantly, AraBART has achieved an accuracy of 94.71%, approaching the performance of Farasa (97.32%). Furthermore, incorporating parts of the Farasa corpus into our training process showed a clear improvement in accuracy, revealing AraBART's effectiveness for broad applications in Arabic NLP tasks, including text summarization and machine translation.
The emergence of a distinct Gen-Z sociolect, often termed "Genzie" or "Internet Slang," represents one of the most rapid and transformative linguistic developments of the digital age. This language is not a random collection of slang but a complex, rule-governed system born from the intersection of technology, social change, and identity formation. A comprehensive, data-driven analysis of its historical evolution, structural properties, and socio-pragmatic functions is critical to understanding contemporary communication, as it reflects fundamental shifts in how a generation conceptualizes interaction, community, and self-expression. This study aims to deconstruct the Gen-Z sociolect by tracing its historical development over a key 36-month period (2021-2023), analyzing its core structural components (lexical, semantic, syntactic, multimodal), and explaining its social functions within digital communities. The research seeks to move beyond anecdotal description to provide a rigorous, empirical account of this dynamic linguistic phenomenon, thereby establishing a benchmark for the academic study of internet-native dialects. We position this sociolect not as a degradation of Standard English, but as a legitimate linguistic innovation worthy of serious scholarly attention, with its own internal logic and systemic coherence. A mixed-methods, diachronic approach was employed, integrating the scale of computational linguistics with the nuance of qualitative discourse analysis. A large-scale corpus of approximately 6000 posts was compiled from three core platforms—Twitter/X, Instagram, and TikTok—across the 12-month timeframe, ensuring a representative sample of public-facing Gen-Z communication. Computational linguistics methods were used for quantitative analysis, including time-series modeling for lexical diffusion, diachronic word embeddings for semantic shift, and supervised machine learning for stylometric identification. This was complemented by qualitative discourse and pragmatic analysis of a stratified sample of posts to understand language-in-use, focusing on the interplay between text, image, and platform-specific conventions. The analysis reveals a clear, platform-influenced historical trajectory for Gen-Z language, with terms originating on niche, visually-driven forums like TikTok and Twitch before achieving mass diffusion on the text-centric environment of Twitter and finally being normalized on the broader social canvas of Instagram. We identified and modeled three primary mechanisms of lexical creation: neologism (e.g., "skibidi," "gyatt"), semantic reappropriation (e.g., "cap," "based," "fire"), and phono-semantic matching from online cultures (e.g., "ratio," "L + RIP bozo"). Gen-Z language is a legitimate and sophisticated dialect of the digital era, a natural linguistic adaptation to a hyper-connected, attention-economy-driven world. Its evolution is not chaotic but follows predictable patterns of cultural transmission that are dramatically amplified and accelerated by social media algorithms. Its structure efficiently manages cognitive load in fast-paced digital environments while its primary functions are the performance of a specific digital identity, the creation and policing of digital community boundaries, and a form of resistance to traditional linguistic and social norms.
This study explores the pragmatic typology of the functional-semantic field (FSF) of degree in English and Uzbek, focusing on how gradability, intensity, and comparison are expressed and interpreted across two typologically different languages. The concept of degree is treated as a universal semantic category realized through a range of linguistic means, including morphological forms, lexical items, and syntactic constructions. The research aims to identify both common patterns and language-specific features in the expression of degree, as well as to analyze the role of pragmatic factors in shaping its meaning. The findings demonstrate that English primarily relies on analytic and morphological devices, such as comparative and superlative forms and intensifiers, while Uzbek employs agglutinative mechanisms, lexical markers, and expressive forms such as reduplication. Despite these structural differences, both languages share a common semantic core based on scalarity and gradation. However, the interpretation of degree is highly context-dependent and influenced by speaker intention, discourse context, and cultural norms. The study also shows that degree expressions serve not only as markers of quantitative or qualitative comparison but also as pragmatic tools for expressing evaluation, emphasis, politeness, and implicature. The functional-semantic field of degree is organized into core and peripheral zones, where core elements provide basic gradation and peripheral elements introduce stylistic and contextual variation. Uzbek demonstrates a stronger tendency toward expressive and emphatic usage, while English often relies on more implicit and context-driven strategies. In conclusion, the research highlights the importance of integrating semantic and pragmatic approaches in the study of degree and contributes to a deeper understanding of cross-linguistic variation in functional-semantic categories. The results have practical implications for language teaching, translation, and intercultural communication.
Klasifikasi sentimen teks menghadapi kompleksitas semantik dan ambiguitas konteks. BERT menawarkan representasi kontekstual dua arah, tetapi strategi adaptasi antara feature extraction dan end-to-end fine-tuning memerlukan pembuktian empiris dari sisi akurasi dan efisiensi komputasi. Penelitian ini menganalisis Model A (feature extraction) dan Model B (end-to-end fine-tuning) berbasis bert-base-uncased pada Stanford Sentiment Treebank (SST-2) dengan 872 sampel held-out test set. Eksperimen dijalankan pada enam random seed dan dievaluasi menggunakan akurasi, precision, recall, F1-score, uji McNemar, Wilcoxon Signed-Rank, Bootstrap 95% Confidence Interval, Cohen's d, konsumsi VRAM, latensi inferensi, dan ketahanan linguistik. Model B mencapai akurasi 92,78% ± 0,71% dan F1-score 92,94% ± 0,79%, sedangkan Model A mencapai 86,05% ± 0,69% dan 86,66% ± 0,43%. Perbedaan signifikan pada Wilcoxon (p=0,03125) dan McNemar (χ²=38,1635; p=6,51×10⁻¹⁰), dengan Cohen's d=10,11. Latensi GPU hampir identik, sedangkan Model A lebih hemat VRAM hingga 68,7% pada clean state. Model B juga lebih kuat pada ulasan panjang dan negasi biner. Hasil ini menunjukkan fine-tuning lebih efektif untuk akurasi dan pemahaman konteks, sementara feature extraction lebih sesuai pada lingkungan dengan keterbatasan memori GPU
Swallowing is tightly coupled with emotion-autonomic states, while olfactory pathways project to limbic and hypothalamic hubs that can enhance parasympathetic activity and reduce stress. Whether a pleasant natural-wood odor can modulate swallowing-related brain activity and behavior remains unclear. To test whether Hinoki (Japanese cypress) inhalation improves affective ratings and increases the repetitive saliva swallowing test (RSST) difference (post-pre; swallows/30 s), and whether it suppresses swallowing-related activity in the premotor cortex (Brodman area 6 (BA6)) and the hypothalamus. The effects of exploratory whole-brain and interaction between Task and Odor were also evaluated. In a randomized within-subject crossover with a one-week interval, participants inhaled Hinoki essential oil or odorless rice-oil control. We assessed brain activity during voluntary saliva swallowing using functional magnetic resonance imaging (fMRI). Outcomes included subjective ratings (odor intensity, preference, comfort, recollection) and RSST counts per 30 s; RSST difference was computed as post-pre inhalation. Hinoki improved affective ratings and increased RSST difference compared to control. The odor main effect showed reduced activation in three regions-bilateral BA6 and the hypothalamus. An interaction effect between Task (swallowing vs. rest) and Odor (Hinoki vs. control) was found in the right putamen. The pattern supports a model: odor-induced autonomic calming (lower heart rate, reduced hypothalamic activity) lowers premotor load and improves swallowing, reflected in RSST difference. These findings suggest that aromatherapy-assisted dysphagia rehabilitation using olfactory priming-particularly for patients with post-stroke sequelae or post-stroke depression with poor appetite-may help enhance training efficiency and recovery.
Social anxiety is marked by persistent fear responses to social cues, often maintained in the absence of repeated aversive social experiences. One potential mechanism underlying this persistence is the recurrent generation of negative social imagery, which may function as an internal source of threat. The present study investigated whether fear toward neutral social stimuli can be acquired through imagined fear-relevant social scenarios alone, and whether this process is heightened in individuals with elevated social anxiety. Using an imagery-based differential conditioning paradigm, neutral facial expressions were paired with fear-relevant social imagery, neutral imagery, or no imagery in individuals with high and low social anxiety. During acquisition, faces associated with fear imagery elicited more negative valence and higher arousal ratings than control stimuli, with these effects amplified in the high social anxiety group. Critically, only individuals with high social anxiety showed reliable differential autonomic responses, including increased skin conductance and heart rate acceleration, to the fear-conditioned stimulus. During extinction, subjective fear and residual autonomic arousal persisted more strongly in the high social anxiety group, indicating impaired safety learning. Together, these findings provide psychophysiological evidence that internally generated threat representations can serve as effective unconditioned stimuli in fear learning, offering a mechanistic account of how negative social imagery contributes to the maintenance of social anxiety and highlighting the potential relevance of interventions targeting maladaptive social imagery.
Abstract - The rapid expansion of digital content on the World Wide Web has made efficient information retrieval an increasingly complex challenge. To assist users in navigating this vast information landscape, several retrieval mechanisms have been developed. Among these, two broad strategies are commonly employed: structured index-based search and autonomous agent-driven search. This work presents an intelligent search agent that leverages a Genetic Algorithm (GA) to perform global web searches. The agent is implemented using the Java programming language and operates by evaluating multiple candidate documents to identify the most relevant result for a given input. The proposed system is built on the Java platform and integrates the Merriam-Webster lexical database to expand query terms with their semantic equivalents. Query words are tokenized, synonym tables are constructed, and varied term combinations are assembled and dispatched to a search engine parser. The returned web pages represent the most semantically aligned results for the original query. Key Words: Genetic Algorithm, web page retrival, searching.
Abstract Pāli exemplifies a less-resourced language: despite a substantial textual corpus, computational infrastructure has lagged behind better-resourced classical languages like Sanskrit, Latin, and Ancient Greek. We present a lemmatized corpus of the Pāli Canon (Tipiṭaka), the canonical scripture of Theravāda Buddhism, a 1.6 million token corpus achieving 99.78% token-level and 97.8% unique-word lemmatization coverage through integration with the Digital Pāli Dictionary. Error analysis of unresolved forms reveals predictable categories: approximately half are potential dictionary additions, with sandhi compounds and metrical spelling variants comprising most of the remainder. The corpus enables previously infeasible computational tasks for Pāli: unsupervised text clustering and topic modeling on normalized vocabulary, training supervised morphological disambiguators, building syntactic treebanks following the Universal Dependencies framework, and conducting cross-lingual alignment with Sanskrit parallel texts. Unlike existing digital editions, this resource provides word-level lemmatization with sandhi decomposition for 42,449 compound forms, specialized handling for titles and metrical variants, and a critical apparatus recording textual variants across five independent witnesses representing the European critical, Burmese, Mahāsaṅgīti, Sinhalese, and Royal Thai editorial traditions. We compare coverage and design decisions against the Digital Corpus of Sanskrit, noting both similarities in scale and important differences in annotation depth and validation methodology. The corpus and reproducible pipeline are released under open licenses.
This study addresses a fundamental question in music psychology: which specific, dynamic acoustic features predict human listeners' emotional responses along the dimensions of valence and arousal. Our primary objective was to develop and validate an interpretable computational model that can serve as a tool for testing and advancing theories of music cognition. Using the publicly available DEAM dataset, containing 1,802 music excerpts with continuous valence-arousal ratings, we developed a novel, theory-guided neural network. This proposed model integrates a convolutional pathway for local spectral analysis with a Transformer pathway for capturing long-range temporal dependencies. Critically, its learning process is constrained by established principles from music psychology to enhance its plausibility. A core finding from an analysis of the model's attention mechanisms was that distinct acoustic patterns drive the two emotional dimensions: rhythmic regularity and spectral flux emerged as strong predictors of arousal, whereas harmonic complexity and musical mode were key predictors of valence. To validate our analytical tool, we confirmed that the model significantly outperformed standard baselines in predictive accuracy, achieving a Concordance Correlation Coefficient (CCC) of 0.67 for valence and 0.73 for arousal. Furthermore, an ablation study demonstrated that the theory-guided constraints were essential for this superior performance. Together, these findings provide robust computational evidence for the distinct roles of temporal and spectral features in shaping emotional perception. This work demonstrates the utility of interpretable machine learning as a powerful methodology for testing and refining psychological theories of music and emotion.
This article focuses primarily on the specific features of Nil Sorsky’s editing of the vocabulary of the Life of Theodore the Studite, which belongs to a distinct corpus of translations from Greek produced in medieval Rus’ by scribes of South Slavic origin. It describes newly identified cases in which semantic archaisms were replaced with equivalents regularly used in Old Russian written tradition (e.g. (домъ → храмъ, достоиныи → любьзныи, искоушенїе → искоусство, мьзда →приѡбрѣтенїе, пытати → испытовати, чинъ → санъ), as well as cases where vernacular Rus- sianisms were substituted with neutral Church Slavonic forms (e.g. потъсноутисѧ → попещисѧ). In addition, drawing on new evidence and the author’s previous research, the article provides a general characterisation of Nil’s system of lexical editing in the antigraphs of four vitae included in the Sobornik. It is established that the principal editorial techniques developed by the elder, under the infl uence of Athonite-Tarnovo traditions and the linguistic norms of original Old Russian hagiographic texts, include: the reduction of features of the local recension of Church Slavonic; the replacement of lexical and semantic archaisms with widely used equivalents in 15th-century literary Church Slavonic texts; the correction of translators’ and scribes’ errors on the basis of context; revisions reflecting the semantic evolution of lexemes; and, fi nally, the selection of synonyms regularly attested in Church Slavonic texts of the Old Russian recension of the late 15th – early 16th century.
Gender-biased communication is pervasive in digital environments. since it can reproduce and amplify socially established beliefs about women and men. This chapter presents an exploratory analysis of gender-associated English vocabulary by combining lexical data from the online Cambridge Thesaurus with computational measures of sentiment classification and online prominence. A set of 47 terms was obtained from the thesaurus entries for manly, masculine, womanlike, and feminine: 29 terms in the masculine-associated set and 18 in the feminine-associated set. Each isolated term was classified with DistilBERT, fine-tuned on the Stanford Sentiment Treebank 2, while approximate Google result totals were collected through SerpApi as a proxy for online prominence. The masculine-associated set contained a larger proportion of negative classifications (31.0%) than the feminine-associated set (22.2%), although the negatively classified feminine-associated terms received higher and less dispersed model-confidence values. Search-result estimates were highly skewed and strongly affected by the literal nouns male and female. Contrary to the initial expectation that polar or negative items would necessarily be more prominent, signed classifier output correlated positively with search-result rank (pooled Spearman’s rho =.51; rho =.63 after excluding male and female). The results therefore reveal asymmetries in lexical coverage and model behaviour but do not establish the sentiment of online discourse or a causal effect of search algorithms. The chapter argues for a cautious, audit-oriented interpretation of decontextualised sentiment classification and outlines a contextualised design for future research on gendered language online.
Abstract Recent advances in text-to-music (TTM) generation have enabled controllable and expressive music creation using natural language prompts, yet the extent to which these systems faithfully convey intended emotions remains largely underexplored. In this study, we introduce AImoclips, a benchmark designed to evaluate emotion conveyance to human listeners in TTM systems using a dimensional valence-arousal framework. We constructed the dataset with 991 instrumental music clips from six TTM systems prompted with 12 emotion words spanning four valence-arousal quadrants. A total of 111 participants provided 6,162 valence and arousal ratings using a 9-point scale. The results revealed that all systems perform above chance in conveying quadrant-level emotion intent, yet overall accuracy remains limited. Notably, substantial model-dependent biases were present. Commercial systems tended to exhibit positive valence deviations, whereas open-source models more often produced outputs with a negative valence shift, with diverging arousal shifts. Furthermore, overall audio quality (Fréchet Audio Distance (FAD)) correlates with both valence and arousal, while text-audio alignment (Contrastive Language-Audio Pretraining (CLAP) score) primarily relates to valence. These findings highlight underlying challenges in conveying linguistic emotional semantics precisely through musically conveyed affect for future research on controllable and perceptually consistent music generation.
Open-vocabulary image classification can recognize arbitrary textual categories but often fails to capture hierarchical relationships, for example, a Ragdoll cat should also be recognized as both cat and animal. Therefore, we introduce OpenHier, a comprehensive framework for open-vocabulary hierarchical classification that encompasses a hierarchical structure, a dataset, a benchmark, and a model. To capture hierarchical relationships, we construct a systematic real-world hierarchical structure based on the linguistic lexical database WordNet and Vision-Language Large Model. Building upon this structure, we develop a large-scale dataset with 4M annotated images, as well as a benchmark with 50K annotated images. Meanwhile, we design an evaluation pipeline to assess open-vocabulary hierarchical classification performance using several curated metrics. Furthermore, we propose a hierarchical consistency constraint method and multimodal alignment strategy to build a hierarchical classification model. Comprehensive experimental results demonstrate that our model achieves superior performance on both multi-label image classification and hierarchical image classification tasks in the open-vocabulary setting.
Music has a well-documented capacity to influence emotional, physiological, and behavioral states, making it a powerful medium for shaping human experience. A fundamental characteristic underlying many of these effects is predictability: listeners continuously form expectations about upcoming musical events based on learned regularities in musical structure. When music confirms or violates these expectations, it can meaningfully shape the listener’s perceptual and emotional experience. Therefore, variations in melodic predictability, ranging from highly regular, familiar patterns to more unexpected or ambiguous sequences, can be expected to differentially affect listeners’ internal states. The objective of this study is to investigate the influence of melodic predictability on subjective and physiological arousal in healthy older adults. Using a within-subject experimental design, 71 older participants listened to melodies varying in melodic predictability while subjective arousal was assessed using a visual analogue scale (VAS), alongside continuous psychophysiological measures of electrodermal activity (EDA) and heart rate (HR). Linear mixed model analyses indicated small but significant effects on subjective arousal ratings and on EDA, indicating psychological and sympathetic effects of melodic predictability. HR results, however, showed small but not significant effects. These findings help us understand melodic predictability as a characteristic of music in influencing arousal in older adults, which may inform the field of music therapy for clinical populations including older adults with dementia. Further research is needed to explore how musical predictability contributes to arousal regulation as a mechanism for supporting older adults with dementia.
Over-reduplication in bilingual aphasia has received limited attention, particularly among Sundanese–Indonesian speakers. This intrinsic case study examined reduplication patterns in one Sundanese–Indonesian bilingual speaker with mixed aphasia following typhoid meningitis. Data from spontaneous interactions, telephone conversations, WhatsApp exchanges, and structured elicitation tasks yielded 40 reduplication tokens. Each token was classified by morphological structure and relation to expected linguistic forms and communicative contexts. Structural categories comprised full reduplication, full reduplication with affixation, partial reduplication with full affixation, partial reduplication with affixation, and partial reduplication (dwipurwa). Productions were classified as anomalous or deviant; anomalous forms were not ordinarily used by general Sundanese speakers, whereas deviant forms departed from expected morphological, grammatical, semantic, or contextual patterns. The participant produced three full reduplication tokens, three full reduplication with affixation tokens, five partial reduplication with full affixation tokens, 28 partial reduplication with affixation tokens, and one dwipurwa token. Partial reduplication with affixation was most frequent (28/40; 70%). Fourteen tokens were anomalous and 37 deviant; classifications were not mutually exclusive. Although the participant produced repeated morphological material, forms were inconsistently aligned with linguistic norms and communicative expectations, suggesting disruption in morphological selection and monitoring. This study offers a detailed case-based description of over-reduplication in a Sundanese–Indonesian bilingual speaker with mixed aphasia, highlighting how typologically specific reduplication systems may be affected. Findings do not establish specific causal or declarative–procedural mechanisms and should not be generalized broadly. Controlled clinical studies with larger, more diverse samples are required before efficacy-based treatment protocols can be recommended.
The article presents a critical analysis of the curriculum for the discipline “Foreign Oriental Language (Chinese Language)” designed for the bachelor’s degree program No. 60230100 — Philology and Language Teaching (Oriental Languages), implemented at the Tashkent State University of Oriental Studies (2024). The relevance of the study is determined by the growing need to rethink the content of Chinese language education in the context of expanding intercultural contacts and the increasing role of the language as a means of professional, academic, and intercultural interaction. The author aims to determine the extent to which the principles of the cognitive- discursive approach and components of intercultural communication are implemented in the analyzed curriculum, as well as to identify priority areas for its further methodological and content-related improvement. The study employs methods of analysis of normative and methodological documentation, content analysis of instructional materials, and comparison of educational objectives with the declared learning outcomes. The curriculum is generally oriented toward the step-by-step formation of language competencies at levels A1–C1 and includes lexical-grammatical, analytical, and research elements that ensure the development of the main types of speech activity. At the same time, the article demonstrates that these elements are predominantly aimed at developing formal linguistic skills and do not sufficiently contribute to the development of discursive competence, cognitive strategies of meaning comprehension, and cognitive-analytical and interpretative strategies of working with texts. According to the author, the intercultural component of the curriculum is presented fragmentarily and is not formulated as an independent learning objective: there is no systematic work with culturally marked discourses, and skills of comparing communicative norms, reflection, and adaptation to intercultural differences are not developed. In conclusion, the necessity of transitioning from a predominantly language- centered model of instruction to a cognitive-discursive and intercultural paradigm is substantiated, aimed at forming foreign-language consciousness and preparing students for effective communication in a multilingual and multicultural environment