Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
16504 papers
Acceptability judgments are one of the major tools for (psycho)linguists to assess speakers’ preferences for specific utterances in a given language, shedding light on the grammar of the language under study. However, it is well known that factors that are not related to grammaticality, such as frequency of exposure, cognitive constraints, and others, can influence the perceived acceptability of an utterance. We will use the system of wh-interrogatives in French as an example to study the impact of linguistic norms on what is considered “good” French. In three experiments, we show that adult L1 French speakers have internalized the dichotomy between variants that are considered “good French”, according to the norms, and those that are suited to more informal daily life situations. Speakers can express these differences when given the appropriate tools, but not with a unique general acceptability scale. In line with previous work, we argue that acceptability judgments are a useful task, but that they need to be refined to account for sociolinguistic factors that constrain speakers’ assessments (i.e., linguistic norms, but also speaker group and formality of the context of interaction).
Lexical Semantic Change (LSC) provides insight into cultural and social dynamics. Yet, the validity of methods for measuring different kinds of LSC remains unestablished due to the absence of historical benchmark datasets. To address this gap, we propose LSC-Eval, a novel three-stage general-purpose evaluation framework to: (1) develop a scalable methodology for generating synthetic datasets that simulate theory-driven LSC using In-Context Learning and a lexical database; (2) use these datasets to evaluate the sensitivity of computational methods to synthetic change; and (3) assess their suitability for detecting change in specific dimensions and domains. We apply LSC-Eval to simulate changes along the Sentiment, Intensity, and Breadth (SIB) dimensions, as defined in the SIBling framework, using examples from psychology. We then evaluate the ability of selected methods to detect these controlled interventions. Our findings validate the use of synthetic benchmarks, demonstrate that tailored methods effectively detect changes along SIB dimensions, and reveal that a state-of-the-art LSC model faces challenges in detecting affective dimensions of LSC. LSC-Eval offers a valuable tool for dimension- and domain-specific benchmarking of LSC methods, with particular relevance to the social sciences.
This paper presents the structure and principal components of the linguistic resources required for sentiment analysis in the Uzbek language. The research aims to identify and develop effective approaches for constructing a linguistic database - referred to as SentiUzNet - and to establish a foundational sentiment lexicon tailored specifically to the characteristics of the Uzbek language. In particular, the paper discusses key principles for annotating words with sentiment polarity and subjectivity scores, as well as methodological foundations for building a lexicographic database to support automated emotional analysis of texts. A significant part of the research focuses on experimenting with large-scale user-generated content, specifically social media comments written in Uzbek. These datasets were used to train and evaluate sentiment analysis models, thereby allowing an assessment of their performance and practical applicability. The results of this research represent one of the first comprehensive attempts to facilitate automatic sentiment detection in the Uzbek language and are expected to contribute substantially to the advancement of natural language processing technologies in under-resourced linguistic settings.
Knowledge distillation (KD) is a widely adopted technique for compressing large models into smaller, more efficient student models that can be deployed on devices with limited computational resources. Among various KD methods, Relational Knowledge Distillation (RKD) improves student performance by aligning relational structures in the feature space, such as pairwise distances and angles. In this work, we propose Quantum Relational Knowledge Distillation (QRKD), which extends RKD by incorporating quantum relational information. Specifically, we map classical features into a Hilbert space, interpret them as quantum states, and compute quantum kernel values to capture richer inter-sample relationships. These quantum-informed relations are then used to guide the distillation process. We evaluate QRKD on both vision and language tasks, including CNNs on MNIST and CIFAR-10, and GPT-2 on WikiText-2, Penn Treebank, and IMDB. Across all benchmarks, QRKD consistently improves student model performance compared to classical RKD. Importantly, both teacher and student models remain classical and deployable on standard hardware, with quantum computation required only during training. This work presents the first demonstration of quantum-enhanced knowledge distillation in a fully classical deployment setting.
This study reviews the English language test of Singapore’s Primary School Leaving Examination, a high-stakes national assessment taken annually by nearly all primary six students for secondary school placement. Given the test’s importance in shaping students’ academic pathways and recent format changes, it is crucial to evaluate its validity, specifically its ability to provide accurate and fair assessments of students’ English language proficiency and academic readiness. The review outlines the test’s educational and policy context, followed by a description of the latest formats for both the English language and foundation English language versions. The analysis focuses on core dimensions of test validity, including content representativeness, construct validity, criterion-related validity (concurrent and predictive), and reliability (inter-rater reliability and internal consistency). Drawing on official documents and limited empirical studies, the review finds moderate improvements in content representativeness and construct validity. However, both longstanding and emerging concerns (e.g., the exclusion of local linguistic norms and genre scope) indicate that key limitations remain. While predictive validity, inter-rater reliability, and internal consistency appear supported, empirical research remains sparse across all reviewed test qualities, particularly in concurrent validity. The review integrates identified research gaps and proposes inquiry directions to inform future test development and policy adaptation. Strengthening the evidence base is essential for ensuring a valid, reliable, and equitable assessment system in Singapore’s primary education landscape.
Abstract This study investigates the relationship between literature and poetry reading frequency and participants’ ratings of metaphors on key features: quality, aptness, familiarity, and comprehensibility. Using a set of Serbian poetic metaphors, we explored two main questions: how reading habits correlate with metaphor feature ratings, and whether the type of reading material (i.e., literature vs poetry) influences sensitivity to these features. The sample consisted of 140 native Serbian-speaking students from varied academic disciplines. Participants rated metaphors based on reading frequency (literature and poetry) using a 7-point Likert scale. Analysis showed that frequent readers generally gave higher overall metaphor ratings than infrequent readers, with significant differences noted particularly in familiarity and comprehensibility. Specifically, familiarity ratings yielded the most substantial differences between infrequent and frequent readers, which can indicate the influence of reading experience on the perceived recognition and understanding of metaphors. Aptness and quality ratings showed no significant differences, which suggests that familiarity and comprehensibility are more sensitive to variations in reading habits.
Abstract Standardized quantitative measurement of texts lies at the heart of digital approaches to humanities. Structure-based textual measures are known to be influenced by the choice of syntactic annotation schemes. Building on previous research, the present article further explores the relation between annotation schemes and the index of mean dependency distance (MDD) by comparing the treebanks of seventeen languages, respectively, within a tree representation (basic universal dependencies, BUD) and within a graphic representation (enhanced universal dependencies, EUD). Following the idea of decomposing annotation schemes into the combinations of analyses of specific constructions (coordinate structures, control constructions, and relative clauses), we design algorithms to identify them in the CoNLL-U format treebanks and explore their influences. It is found that the overall MDD of the EUD representation is statistically higher than that of BUD at corpus level, primarily affected by the coordinate structure due to its high frequency. At sentence level, all three constructions might contribute to either increased or decreased MDD, with stochastically intervening words and word order being two important determinants of the values of the measure. Finally, we propose and argue for the view that MDDs calculated under different annotation schemes should be regarded as different textual measures in nature. In sum, the present study provides another case study to deepen our understanding of the nature of syntactic annotation schemes and its relation with textual indices, which paves the way for standard measurement of texts in future humanities research.
Introduction Researchers working in the field of cognitive aging frequently encounter highly motivated yet nervous older participants during data collection in the laboratory. Such anecdotal experiences raise the question of whether the affective or physiological response of older participants to psychological laboratory experiments differs to that of young adults, who might be less motivated but also less nervous, as they may be more used to the environment and to learning and memory tests. Methods In the present study, we collected saliva samples and subjective affective ratings during an EEG experiment on memory, and at home, in young and older adults, while also taking into account sex effects. Results There was no significant interaction involving time point (laboratory vs. at home) and age group. However, across both time points older males showed significantly higher cortisol-levels than older females, while there was no difference for younger males and females. The trajectories in cortisol levels throughout the session, especially around the memory task, differed by age: While there was a decrease in cortisol levels for younger adults from before to after the memory task, we did not observe such a decrease in older participants. There were few age differences in alpha-amylase or negative affect. However, older adults showed higher ratings of positive affect than younger participants. Importantly, lower cortisol levels before the memory task were associated with higher associative memory performance for older adults. Discussion Affective reactions to psychological laboratory tasks may hence be an important factor to consider in psychological experiments in the field of cognitive aging.
Decision confidence is a prototypical metacognitive representation that is thought to approximate the probability that a decision is correct. The perception of being correct has also been associated with affective valence such that being correct feels more positive and being mistaken more negative. This suggests that, similarly to confidence, affective valence reflects the probability that a decision is correct. However, both fields of research have seen very little interaction. Here, we test if affect, similarly to confidence reflects probability that a decision is correct in two perceptual decision-making experiments where we compare the relationships of theoretically relevant variables (e.g. evidence, accuracy, and expectancy) with both confidence and affect ratings. The findings indicate that confidence and affect ratings are similarly sensitive to changes in accuracy, evidence, and expectancy, indicating that both track the subjective probability that a decision is correct. We identify various mechanisms that can explain these results. We also envision future research for clarifying the role of cognitive and affective aspects of metacognition relying on deeper integration of the respective research fields.
Emotional experiences involve dynamic multisensory perception, yet most EEG research uses unimodal stimuli such as naturalistic scene photographs. Recent research suggests that realistic emotional videos reliably reduce the amplitude of a steady-state visual evoked potential (ssVEP) elicited by a flickering border. Here, we examine the extent to which this video-ssVEP measure compares with the well-established Late Positive Potential (LPP) that is reliably larger for emotional relative to neutral scenes. To address this question, 45 participants viewed 90 matched pairs of realistic videos and scenes. Consistent with prior work, reduced 7-8 Hz ssVEP amplitude was evident during emotional relative to neutral videos. However, this reduction in power was not specific to the driving frequency of 7.5 Hz, and in fact, Fourier transformation analyses limited to 7.5 Hz were not modulated by video content. Still, at the group level, the video-driven reductions in 7-8 Hz power and LPP modulation by scenes produced similarly large valence effects, and both measures strongly correlated with arousal ratings. Consistent with previous research, the scene-LPP was sensitive to specific emotional contents (erotica and gore) somewhat inconsistent arousal ratings. In contrast, the video-driven oscillation modulation did not show this content sensitivity and was better explained by individual arousal ratings per video clip. In sum, these results show that the 7.5 Hz flickering-border paradigm does not index emotional engagement with video stimuli, yet emotional videos do evoke robust decreases in 3-10 Hz oscillatory power that is somewhat distinct from emotional modulation of the scene-evoked LPP. Matched emotional video and scenes evoke large EEG responses compared with neutral content within-participant. Our findings align with previous research indicating that video modulation of power around the evoked 7.5 Hz ssVEP frequency (7-8 Hz) serves as a reliable emotional measure. However, further analyses reveal that this effect is attributable to a general decrease in power across the 3-10 Hz frequency range.
Over the past two decades, quantitative models and statistical methods have been increasingly applied to syntax research, with treebanks emerging as a vital resource for this field. However, tools specifically designed for calculating syntactic metrics remain relatively scarce compared to traditional word frequency measures. This gap hinders the deeper integration of quantitative methods into syntax research. In response, we have developed a tool called QuanSyn, which is tailored for quantitative syntax analysis. Based on treebanks, QuanSyn can calculate metrics such as dependency distance, hierarchical distance, dependency direction, and valency-related syntactic metrics. Additionally, it facilitates the construction of linguistic networks and rapid fitting of functions. QuanSyn can help lower the barriers to accessing syntactic quantitative analysis, thereby promoting the broader application of quantitative methods in syntax research.
Detection of out-of-distribution (OOD) samples is cru-cial for safe real-world deployment of machine learning models. Recent advances in vision language foundation models have made them capable of detecting OOD sam-ples without requiring in-distribution (ID) images. How-ever, these zero-shot methods often underperform as they do not adequately consider ID class likelihoods in their detection confidence scoring. Hence, we introduce CLIPScope, a zero-shot OOD detection approach that normalizes the confidence score of a sample by class likelihoods, akin to a Bayesian posterior update. Furthermore, CLIPScope incor-porates a novel strategy to mine OOD classes from a large lexical database. It selects class labels that are farthest and nearest to ID classes in terms of CLIP embedding distance to maximize coverage of OOD samples. We conduct ex-tensive ablation studies and empirical evaluations, demon-strating state of the art performance of CLIPScope across various OOD detection benchmarks. Code is available at https://github.com/ful001hao/CLIPScope.
Formulaic Networks as Prototypical Categories: Combining the Ancient Greek Dependency Treebank with the Ancient Greek WordNet for a Pilot Study on the Iliad was published in Advances in Ancient Greek Linguistics on page 737.
The linguistic features of the Uzbek language - complex agglutinative morphology, free word order, and limited resources - necessitate a specialized approach and thorough research in the application of morphological and syntactic methods. Within the framework of the study, morphological analysis methods and syntactic analysis methods are reviewed based on scientific sources. Each section presents the existing advantages and disadvantages, experience of their use in the Uzbek language, as well as a comparative analysis with foreign languages. Rule-based methods, statistical models (HMM, CRF, etc.), Neural network-based approaches (BiLSTM-CRF, seq2seq) of morphological analysis in the Uzbek language are discussed, and the results are given in examples and percentages. It is shown that syntactic parsing is implemented using dependency and constituency parsing analysis methods. The issue of building a UD treebank for the Uzbek language with SOV order is considered. The impact of complex morphological structure and free word order in sentences on the construction of parsers is highlighted. As a result of the studied approaches, the issue of building hybrid parsers, integrating them with morphological analysis and assigning grammatical categories of words to the parser is raised. Also, the development of neural constituency parsers based on neural networks and the effectiveness of the results obtained from them are analyzed.
Background and aims: Despite a previously reported connection between compulsive sexual behaviors (CSB), such as problematic pornography use, and heightened cue-reactivity, empirical evidence of the alteration of processes responsible for increased salience attribution to erotic cues remains sparse. Drawing on similarities with addiction models, this study explores the neuronal mechanisms of CSB through the use of appetitive conditioning and extinction with erotic and monetary rewards. Methods: Thirty-two heterosexual males struggling with CSB (age: 28.9 ± 7.1), and 31 healthy matched participants (age: 27.8 ± 5.6) underwent active appetitive conditioning and extinction tasks in fMRI. The effects of conditioning and extinction towards cues of erotic and monetary rewards were measured via self-assessment (valence and arousal rating towards cues), behavior (reaction times), and brain reactivity. Results: In conditioning, subjective ratings increased, and reaction times were faster for both erotic and monetary cues among participants with CSB, along with altered activity in ventral striatum (vStr), dorsal anterior cingulate cortex (dACC), and anterior orbitofrontal cortex (aOFC). In extinction, self-assessment ratings remained elevated in the CSB group for both cues in a non-reward-specific fashion, accompanied by altered activity of dACC and vStr. Discussion and conclusions: These findings suggest enhanced incentive salience attribution to conditioned cues, highlighting a generalized motivational and value-related transfer from rewards to the cues in participants with CSB. Additionally, despite the absence of rewards, the persistence of arousal and valence towards cues underscored the maladaptive extinction process. These insights advance the understanding of CSB's neurobiological underpinnings and its relation to addiction frameworks.
<span lang="EN-US">The illicit act of appropriating programming code has long been an appealing notion due to the immediate time and effort savings it affords perpetrators. However, it is universally acknowledged that concerted efforts are imperative to identify and rectify such transgressions. This is particularly crucial as academic institutions, including universities, may inadvertently confer degrees for work tainted by this form of plagiarism. Consequently, the primary objective of this research is to scrutinize the feasibility of identifying plagiarism within pairs of Verilog algorithms and texts. this study aims to detect plagiarism in textual content and Verilog code by leveraging diverse linguistic characteristics from the WordNet lexical database. The primary objective is to achieve optimal accuracy in identifying instances of plagiarism, incorporating features such as modifications to text structure, synonym substitution, and simultaneous application of these strategies. The system's architecture is intricately designed to unveil instances of plagiarism in both textual content and Verilog code by extracting nuanced characteristics. The systematic process includes preprocessing, detailed analysis, and post-processing, supported by a feature-rich database. Each entry in the database represents a distinctive similarity case, contributing to a thorough and comprehensive approach to plagiarism detection.</span>
Emojis are widely used in digital communication to convey emotional cues alongside text, yet their impact on word-level reading within sentence contexts remains unclear. We conducted an eye-tracking experiment to examine how positive (e.g., 🤩) versus neutral (e.g., 🧑🦳) face emojis embedded mid-sentence in otherwise neutral sentences affect the processing of the preceding and following words (e.g., positive “Did you change your hair 🤩 something is different” vs. neutral “Did you change your hair 🧑🦳 something is different”). We observed robust parafoveal-on-foveal (PoF) effects on the n–1 word, with longer fixations in first-fixation, gaze duration, and single-fixation measures when the parafoveal emoji was positive rather than neutral. This valence effect persisted even after accounting for mislocated fixations, suggesting that positive emotional content genuinely modulates foveal word processing. In contrast, the n+1 word showed no valence-based facilitation, implying that the influence of a positive mid-sentence emoji does not extend to subsequent words in continuous reading. At the sentence level, positive emojis were associated with faster overall reading times and higher valence ratings, although dashed (no-emoji) sentences in the pre-test were rated more positively than emojified versions in the experiment. These findings reinforce models of eye movement control that allow parallel processing of foveal and parafoveal information, highlighting how affective face emojis can shape real-time reading dynamics.
BACKGROUND: Emotion dysregulation is a central feature in trauma-associated disorders such as posttraumatic stress disorder (PTSD) and borderline personality disorder (BPD). However, it remains unclear whether emotion dysregulation is a transdiagnostic phenomenon closely linked to childhood trauma, or if disorder-specific alterations in emotion processing exist. Following a multimethodological approach, we aimed to assess and compare the reactivity to and regulation of emotions between patients with BPD and PTSD, as well as healthy controls, and identify associations with childhood trauma. METHODS: A total of 135 women, 43 healthy controls, 43 with BPD and 49 with PTSD, took part in a multimethodological assessment of emotional reactivity and regulation. Self-report measures were used to assess childhood trauma and emotion dysregulation. Additionally, participants performed a classic emotion regulation (ER) paradigm. Subjective emotional valence ratings and neurophysiological responses (P3 and late positive potential, LPP) were measured in response to negative, positive, and neutral pictures (emotional reactivity) and during active regulation vs. passive viewing of negative pictures (ER). RESULTS: Regarding emotional reactivity, during the experimental paradigm both patient groups reported lower emotional valence after viewing positive or neutral pictures compared to healthy controls. Furthermore, P3 amplitudes in response to neutral pictures were reduced in both patient groups and in response to negative pictures, specifically in patients with PTSD. Regarding ER, while both patient groups self-reported significant disturbances in ER, neither valence ratings nor neurophysiological responses assessed during the ER task (P3, LPP) differed from healthy controls. Across groups, childhood trauma was related to decreased emotional valence ratings on neutral and positive pictures and higher self-reported emotion dysregulation. CONCLUSIONS: Patients with BPD and PTSD exhibited a reduced emotional reactivity in response to positive and neutral information. Specifically, patients with PTSD demonstrated hypo-reactivity to neutral and trauma-unrelated negative stimuli, which might be due to altered attentional resource allocation following trauma. Although patients reported using adaptive ER strategies less frequently in daily life, they effectively implemented them when instructed to, highlighting important clinical and theoretical implications.
Deep neural networks employ specialized architectures for vision, sequential and language tasks, yet this proliferation obscures their underlying commonalities. We introduce a unified matrix-order framework that casts convolutional, recurrent and self-attention operations as sparse matrix multiplications. Convolution is realized via an upper-triangular weight matrix performing first-order transformations; recurrence emerges from a lower-triangular matrix encoding stepwise updates; attention arises naturally as a third-order tensor factorization. We prove algebraic isomorphism with standard CNN, RNN and Transformer layers under mild assumptions. Empirical evaluations on image classification (MNIST, CIFAR-10/100, Tiny ImageNet), time-series forecasting (ETTh1, Electricity Load Diagrams) and language modeling/classification (AG News, WikiText-2, Penn Treebank) confirm that sparse-matrix formulations match or exceed native model performance while converging in comparable or fewer epochs. By reducing architecture design to sparse pattern selection, our matrix perspective aligns with GPU parallelism and leverages mature algebraic optimization tools. This work establishes a mathematically rigorous substrate for diverse neural architectures and opens avenues for principled, hardware-aware network design.
BACKGROUND AND OBJECTIVES: It is well documented that the fear of specific stimuli and situations can be acquired through the social observation of the actions of another person. In contrast, it is still a matter of debate, whether processes related to fear attenuation, extinction, and extinction-retrieval can equally be achieved through social observation after de novo fear conditioning. METHODS: Here, we used a differential fear conditioning procedure and investigated whether the variation of the context of video-based vicarious extinction learning (VEL) will affect subsequent extinction learning and extinction-retrieval. Conditioned fear acquisition, extinction, and extinction-retrieval was measured using psychophysiological (skin conductance responses) and subjective measures (CS-UCS contingency ratings and CS-valence ratings). RESULTS: Participants showed enhanced fear extinction learning after VEL as compared to controls. VEL improved extinction learning relative to controls but appeared to be highly context-dependent. The beneficial effect of VEL on subsequent extinction learning was abolished when the context in which the model was performing in the video was different from the context in which the observer performed all stages of the experiment. LIMITATIONS: Data were obtained in a non-clinical sample which does not permit the extrapolation of findings to clinical populations. CONCLUSION: Our results suggests that safety information derived from VEL promotes fear extinction when model and observer perform the experiment in the same context. Given that fear extinction is considered as an experimental proxy of exposure therapy, our findings might be instructive for the development of novel clinical interventions to promote exposure treatment efficacy.
This study investigates the relationship between the affective response evoked by hearing upstairs neighbour footsteps in wooden residential buildings, the acoustic characteristics of those footsteps sounds, and the personal traits of participants. Sound recordings were analysed using parameters extracted from the autocorrelation function (ACF) and interaural cross-correlation function (IACF) to identify temporal and spatial features. A laboratory experiment involving 46 individuals assessed their affective responses in terms of arousal and valence after being exposed to a variety of footsteps sounds. The visual simulation of the living room scenario was generated by inviting participants to wear a head-mounted display (HMD) that showed a 360-degree image of a living room with natural or artificial lighting. Participants self-reported their non-acoustic traits before the test including noise sensitivity, attitude towards neighbours, and circadian rhythm type. Results showed significant correlations between affective responses and all acoustic parameters. Pitch-related parameters (ɸ 1 and τ 1 ) and sound pressure level (SPL) were good predictors of arousal and valence. Without SPL, spatial parameters (IACC and τ IACC ) also contributed to affective ratings. Furthermore, participants with low noise sensitivity and a Morning chronotype reported significantly lower arousal and higher valence compared to those with high noise sensitivity and an Evening chronotype. Finally, participants with a positive attitude towards neighbours exhibited higher valence than those with a negative attitude towards neighbours. This study uniquely explores emotional responses to neighbour sounds in lightweight wooden buildings integrating acoustic and non-acoustic factors and using immersive simulations for a more realistic and ecologically valid assessment. • Arousal and valence ratings were significantly correlated with SPL and parameters extracted from the ACF/IACF functions. • Pitch-related parameters and SPL were good predictors of arousal and valence. • Without SPL, spatial parameters also contributed to affective ratings. • Non-acoustic factors significantly impacted affective responses and their correlations with ACF/IACF parameters.
This study investigated the neurophysiological and affective responses elicited by nature-inspired indoor design elements, including curvilinear forms (CL), nature views (N), and wooden interiors (W), in a virtual environment, and their effects on cognitive performance. Thirty-six participants experienced one control and three experimental conditions in a within-subject design. Electroencephalography (EEG) was used to record neural activity, relaxation and valence ratings assessed affective states, and standardized tasks measured cognitive performance. The W condition elicited EEG patterns indicative of relaxed attentional engagement, including increased alpha-to-theta (ATR) and alpha-to-beta (ABR) ratios, and a decreased theta-to-beta (TBR) ratio. These neural patterns were associated with higher self-reported relaxation and positive affect, and with enhanced cognitive performance relative to the control condition. In contrast, the CL and N conditions did not improve cognitive performance, and the N condition showed elevated physiological arousal, likely due to heightened visual stimulation. Regression analysis identified ATR and relaxation as significant predictors of cognitive performance, emphasizing the role of emotional stability and neural balance in supporting task engagement. Overall, the findings highlight the potential of nature-inspired design to foster a synergy between psychological relaxation and cognitive attention, though further research is needed across diverse spatial typologies to isolate specific design parameters.
This study explores the syntactic network characteristics of English e-commerce live-streaming discourse by employing a syntactic treebank and syntactic complex network analysis. The main findings are: (1) The syntactic network of English e-commerce live-streaming discourse exhibits small-world and scale-free properties, which are hallmark traits of complex networks. (2) The central nodes of the network are be, I, and the, with be serving as the most central node, while I and the act as local central nodes. (3) The central node be demonstrates both strong centrifugal and centripetal forces. Its centrifugal force is most frequently associated with subject relations and adjective complements, while its centripetal force is characterized by auxiliary and clausal complements. These findings indicate that the syntactic structure of English e-commerce live-streaming discourse is highly robust. This robustness underscores the discourse’s functional purpose: to convey information clearly while engaging users through personalization and specificity. Furthermore, the study highlights the critical role of be in attributive and descriptive constructions. Overall, this research provides insights into the syntactic organization of e-commerce discourse and demonstrates the effectiveness of complex network analysis in linguistic studies.
The quantity and variety of Old Irish text which survives in contemporary manuscripts, those dating from the Old Irish period, is quite small by comparison to what is available for Modern Irish, not to mention better-resourced modern languages.As no native speakers have existed for more than a millennium, no more text will ever be created by native speakers.For these reasons, text surviving in contemporary sources is particularly valuable.Ideally, all such text would be annotated using a single, common standard to ensure compatibility.At present, discrete Old Irish text repositories make use of incompatible annotation styles, few of which are utilised by text resources for other languages.This limits the potential for using text from more than any one resource simultaneously in NLP applications, or as a basis for creating further resources.This paper describes the production of the first Old Irish text resources to be designed specifically to ensure lexical compatibility and interoperability.
The present study extends recent work on Universal Dependencies annotations for secondlanguage (L2) Korean by introducing a semiautomated framework that identifies morphosyntactic constructions from XPOS sequences and aligns those constructions with corresponding UPOS categories.We also broaden the existing L2-Korean corpus by annotating 2,998 new sentences from argumentative essays.To evaluate the impact of XPOS-UPOS alignments, we fine-tune L2-Korean morphosyntactic analysis models on datasets both with and without these alignments, using two NLP toolkits.Our results indicate that the aligned dataset not only improves consistency across annotation layers but also enhances morphosyntactic tagging and dependency-parsing accuracy, particularly in cases of limited annotated data.
This manuscript proposes the S M Nazmuz Sakib Dependency-Focus Principle for Bengali sentence structure and introduces a derived scalar quantity, the Sakib constant, defined over dependency treebanks. Informally, the principle states that in attested Bengali usage, core arguments (subjects and objects) cluster closer to the verbal head than peripheral modifiers (adverbials and clausal adjuncts), and that the ratio between these average distances is numerically stable across corpora. Using real statistics from the UD Bengali-BRU treebank and the Bengali section of the Bengali-Magahi PUD treebank, we define the Sakib constant K Sakib as the ratio between average dependency lengths of core versus peripheral relations, and compute its value for UD Bengali-BRU. Ten figures based on genuine counts and averages illustrate tense and case distributions, relation frequencies, and core versus non-core dependency lengths for Bengali and, for comparison, Magahi. The proposal is presented as a precise hypothesis, mathematically well-defined and empirically grounded in existing treebank data, but still requiring broader testing for confirmation and cross-linguistic generalisation.
We expand the second language (L2) Korean Universal Dependencies (UD) treebank with 5,454 manually annotated sentences. The annotation guidelines are also revised to better align with the UD framework. Using this enhanced treebank, we fine-tune three Korean language models and evaluate their performance on in-domain and out-of-domain L2-Korean datasets. The results show that fine-tuning significantly improves their performance across various metrics, thus highlighting the importance of using well-tailored L2 datasets for fine-tuning first-language-based, general-purpose language models for the morphosyntactic analysis of L2 data.
This repository contains data accompanying the publication "Auditory localization and subjective assessment of autonomous cleaning robot sounds: A VR experiment on speed, operating mode and alerting signals", submitted for review to the Acta Acustica. The dataset contains: Audio and video material Stimuli consisting of robot recordings under all evaluated conditions (0.3 m/s and 0.8 m/s speed, with and without cleaning, with and without noise AVAS or multi-tone AVAS, both with and without added amplitude modulation). All sounds were exported as 32-bit float wav files; i.e., reading the files into Matlab with audioread results in calibrated Pa values. The files uploaded here were used as source signals in the auralization, assuming a distance of 1 m. The final binaural stimuli were rendered by TASCAR and include an attenuation corresponding to the simulated 7 m distance. 30cms_cleaning_noAVAS.wav 30cms_noCleaning_multiTone.wav 30cms_noCleaning_multiToneAM.wav 30cms_noCleaning_noAVAS.wav 30cms_noCleaning_noise.wav 30cms_noCleaning_noiseAM.wav 80cms_cleaning_noAVAS.wav 80cms_noCleaning_multiTone.wav 80cms_noCleaning_multiToneAM.wav 80cms_noCleaning_noAVAS.wav 80cms_noCleaning_noise.wav 80cms_noCleaning_noiseAM.wav ambienceNoise.wav Excerpt of background noise played back during the experiment. localizationTaskDemo.mp4 Participant POV recording of localization task. This recording was done with a fixed head position, in the actual experiment participants were turning their heads freely. Experiment results and analysis localizationData.csv Table containing the mean and standard deviation of absolute localization error, aggregated for each participant and stimulus. subjectiveData.csv Table containing mean and z-scored annoyance, arousal, trust, and valence ratings for each participant and stimulus. stimuliAnalysis.csv Table containing results of level, loudness, sharpness, roughness, tonality, fluctuation strength, and impulsiveness analysis for all stimuli.
Peer reviewed: True
Stella Markantonatou, Vivian Stamou, Stavros Bompolas, Katerina Anastasopoulou, Irianna Linardaki Vasileiadi, Konstantinos Diamantopoulos, Yannis Kazos, Antonios Anastasopoulos. Proceedings of the 21st Workshop on Multiword Expressions (MWE 2025). 2025.
Cette thèse présente le corpus NaijaSynCor-Prosody, une ressource innovante qui intègre des annotations phonétiques détaillées dans un corpus syntaxique existant. Chaque token du corpus est associée à des annotations décrivant la hauteur, la durée, l’intensité et d’autres attributs prosodiques de chaque syllabe. Ces annotations incluent notamment des contours prosodiques stylisés produits à l’aide du modèle SLAM 3, développé au cours de cette thèse. Cette fusion de données morphosyntaxiques et prosodiques permet des analyses quantitatives de l’interaction entre intonation et syntaxe dans le Naijá, ou pidgin nigérian. À l’aide de ce corpus, nous étudions la différenciation prosodique des unités lexicales qui remplissent également une fonction grammaticale, par exemple des auxiliaires préverbaux marqueurs de TAM. Les contrastes prosodiques entre usages lexicaux et grammaticaux ne sont pas uniformes selon les éléments, bien que notre corpus révèle deux grandes classes prosodiques d’auxiliaires. Un groupe (dey, go, bin) présente une hauteur faible et une courte durée, tandis que les autres se caractérisent globalement par une hauteur plus élevée et une durée plus longue. Deux d’entre eux, make et no, sont particuliers car ils combinent une hauteur élevée et une faible durée. Une analyse plus approfondie de leurs environnements syntaxiques montre qu’ils ont également des distributions atypiques parmi les auxiliaires. Par ailleurs, plusieurs adverbes annotés présentent une distribution syntaxique analogue à celle des auxiliaires et un profil prosodique similaire aux auxiliaires à haute hauteur et longue durée. Ce travail soulève des questions plus générales sur l’utilisation des annotations prosodiques pour remettre en question les étiquettes morphosyntaxiques existantes. La thèse fournit ainsi à la fois une nouvelle ressource linguistique et un cadre méthodologique pour intégrer l’analyse prosodique dans l’étude syntaxe–prosodie à partir de corpus.
Treebanks are critical resources in Natural Language Processing (NLP), supporting parser development, linguistic research, and the evaluation of large language models. While Tamil has seen progress in Universal Dependencies (UD) treebanking, existing corpora have been restricted to prose texts, leaving its vast poetic tradition underrepresented. This paper presents the first effort to be made to construct a syntactic treebank for Tamil poetry, specifically focussing on the ThirukkuRaḷ, which is composed in kuRaḷ veṇpā form. A central challenge in this work is posed by the treatment of multiword tokens (MWTs) and elliptical constructions, both of which are observed to occur frequently in Tamil verse due to its agglutinative morphology and metrical constraints. An annotation strategy is proposed within the Enhanced UD (EUD) framework to systematically address five major types of ellipsis—casal, verbal, adjectival, comparative/simile, and cumulative—alongside complex MWT patterns. These annotations not only enhance the representation of Tamil poetic syntax but also broaden the applicability of UD guidelines to underrepresented genres. The contribution is shown to underscore the linguistic and computational importance of capturing the structural specificities of Tamil poetry, while establishing a foundation for future cross-linguistic and literary treebanking efforts.
Background: Many previous studies highlighting a relationship between depression and emotional face recognition have relied on measures of classification accuracy to determine recognition deficits. However, the perception of emotions is also related arousal levels and valence, and more research is needed to determine how depression impacts these dimensions.Aims: To compare performance on both an objective forced choice emotional recognition task and subjective emotional face valence rating task in participants with self-reported high depression.Methods: Based on screening using the depression sub-scale of the DASS-42, 46 participants (23 males, 23 female) were in the high depression group (mean DASS-42 34±5) and 50 participants in the control groups (25 males, 25 females) with DASS-42 scores of either 0 or 1. All participants completed both a performance-based task (objective) as well as a rating task (subjective) of emotional facial expressions. Results: The data indicate that difference in performance exist in classification accuracy between the groups, with depressed participants demonstrating reduced accuracy for anger, sadness and neutral facial expressions. Additionally differences in subjective ratings exist in the depressed group, but with the important caveat that these only relate to faces display positive emotional expressions.Discussion: The limitations of relying solely on objective tasks where recognition accuracy is the main outcome measure are discussed as well as the data quantitatively demonstrating a reduced response in the depression group to positive stimuli. This study justifies the need for future studies using both objective and subjective measures to assess emotion classification deficits in depression.
The increasing availability of cross-linguistic databases dedicated to documenting morphosyntactic, lexical and phonological features has proliferated the use of such data for studies on language evolution and human history. However, most of these databases were not designed to ensure independence of features, such that it is not valid to jointly use all their features in large-scale statistical analyses assuming independence of inputs. Here, we curate published data from five large linguistic databases to generate two global-scale cross-linguistic datasets: GBI (from the Grambank dataset), and TLI (using inputs from the World Atlas of Language Structures, AUTOTYP, PHOIBLE and Lexibank). The datasets minimize logical dependencies of features and forms of strong statistical dependencies that go beyond phylogenetic and geographical signal. They are also made available in densified form, reducing the proportion of missing data. We document our curation principles and workflows to ensure reusability of this framework with other inputs or thresholds of independence. Our curation steps on both datasets reveal robust and comparable global patterns of structural linguistic diversity.
Multilingual Large Language Models (LLMs) have shown remarkable performance across various languages; however, they often include significantly less data for low-resource languages such as Urdu compared to high-resource languages like English. To assess the linguistic knowledge of LLMs in Urdu, we present the Urdu Benchmark of Linguistic Minimal Pairs (UrBLiMP) i.e. pairs of minimally different sentences that contrast in grammatical acceptability. UrBLiMP comprises 5,696 minimal pairs targeting ten core syntactic phenomena, carefully curated using the Urdu Treebank and diverse Urdu text corpora. A human evaluation of UrBLiMP annotations yielded a 96.10% inter-annotator agreement, confirming the reliability of the dataset. We evaluate twenty multilingual LLMs on UrBLiMP, revealing significant variation in performance across linguistic phenomena. While LLaMA-3-70B achieves the highest average accuracy (94.73%), its performance is statistically comparable to other top models such as Gemma-3-27B-PT. These findings highlight both the potential and the limitations of current multilingual LLMs in capturing fine-grained syntactic knowledge in low-resource languages.
Information on the relationship between facial thermal responses and emotional state is valuable for sensing emotion. Yet, previous research has typically relied on linear methods of analysis based on regions of interest (ROIs), which may overlook nonlinear pixel-wise information across the face. To address this limitation, we investigated the use of machine learning (ML) for pixel-level analysis of facial thermal images to estimate dynamic emotional arousal ratings. We collected facial thermal data from 20 participants who viewed five emotion-eliciting films and assessed their dynamic emotional self-reports. Our ML models, including random forest regression, support vector regression, ResNet-18, and ResNet-34, consistently demonstrated superior estimation performance compared to traditional simple or multiple linear regression models for the ROIs. To interpret the nonlinear relationships between facial temperature changes and arousal, saliency maps and integrated gradients were used for the ResNet-34 model. The results show nonlinear associations of arousal ratings in nose = tip, forehead, and cheek temperature changes. These findings imply that ML-based analysis of facial thermal images can estimate emotional arousal more effectively, pointing to potential applications of non-invasive emotion sensing for mental health, education, and human-computer interaction.
Thermal imaging technology, known for its noncontact and noninvasive nature, offers distinct advantages in computerized emotion sensing. In the literature, a decrease in nose-tip temperature has been associated with dynamic subjective arousal. However, these studies were limited by their focus on a few regions of interest, neglecting a comprehensive analysis of the entire face, and not accounting for the temporal dynamics of thermal changes. To overcome these limitations, we propose an analytical method for facial thermal images using statistical parametric mapping (SPM), which was developed for functional brain image analysis. We developed semiautomated preprocessing protocols to effectively realign and standardize facial thermal images. To validate these analyses, we recorded the thermal images of participants’ faces and assessed dynamic valence and arousal ratings while they observed emotional films. The proposed SPM analyses revealed significant negative associations with dynamic arousal ratings at the nose tip and forehead. The analyses incorporating temporal disparity revealed more forehead clusters than the analyses assuming no delay. These findings validate the proposed pixel-based facial thermal image analysis method using SPM. The results suggest that computerized pixel-based analysis of facial thermal images can be used to estimate dynamic emotional states, with potential applications in various human behavioral fields, including mental health diagnosis and marketing research. • We developed a pixel-based facial thermal image analytical method using SPM. • The method for emotion sensing was tested with an experiment showing emotional films. • Facial thermal images and dynamic valence/arousal ratings were measured. • Negative associations with arousal ratings were detected at the nosetip and forehead. • The data validate the pixel-based facial thermal image analysis for emotion sensing.
Abstract The age-related positivity bias refers to the finding that older adults recount events more positively (or less negatively) as compared to younger adults (i.e., a main effect of age on memory valence). This bias is closely related to the positivity effect, which reflects an interaction between age and valence of information to be remembered. We examined the age-related positivity bias and positivity effect using a one-year longitudinal design with a sample that spanned adulthood (N = 374; age range 19-90; M= 47.41; SD= 16.75). Participants answered questions regarding their memories of learning about the outcome of the 2020 U.S. presidential election. Analyses examined the association between age and valence ratings (positive, negative) and ratings of feelings (happy, elated, upset, and shaken) at Time 1, as well as the association with age between change scores for each of those variables, while controlling for who the participant voted for in the election. Results indicate that increased age was associated with reporting feeling less negative at the time of the event, and also remembering feeling more positive (elated and happy) when reconstructing the event one year later, thereby providing evidence of the positivity bias. There was no evidence of an age by valence interaction in a 2 (Valence) x 3 (Age) mixed ANCOVA on the positive and negative change scores, indicating there was not a positivity effect. Depressive symptoms partially mediated the relationship between age and valence variables, indicating that depressive symptoms may be one mechanism for explaining the age-related positivity bias.
Word Sense Disambiguation (WSD) is a fundamental task in Natural Language Processing (NLP), addressing the challenge of identifying correct word meanings in context. This task is particularly complex for morphologically rich and resource-limited languages like Hindi, which exhibit significant lexical ambiguity compounded by limited availability of annotated corpora. To address these challenges, we propose a supervised approach combining the multilingual BERT model (mBERT) with Hindi WordNet as a structured lexical resource. Using few-shot learning, we fine-tune mBERT on a dataset constructed from Hindi WordNet to disambiguate contextually ambiguous words across four parts of speech (POS): nouns, verbs, adjectives, and adverbs. Experiments on standard Hindi WSD benchmarks demonstrate that our method significantly outperforms traditional rule-based and embedding-based approaches, achieving 96.48% accuracy—an approximate 3% improvement over the strongest baseline. These results validate the effectiveness of integrating contextualized embeddings from pre-trained language models with structured lexical databases, highlighting the promise of hybrid techniques for advancing WSD in low-resource languages and providing a framework applicable to other morphologically complex languages with similar resource constraints.
The nouns of our language refer to either concrete entities (like a table) or abstract concepts (like justice or love), and cognitive psychology has established that concreteness influences how words are processed. Accordingly, understanding how concreteness is represented in our mind and brain is a central question in psychology, neuroscience, and computational linguistics. While the advent of powerful language models has allowed for quantitative inquiries into the nature of semantic representations, it remains largely underexplored how they represent concreteness. Here, we used behavioral judgments to estimate semantic distances implicitly used by humans, for a set of carefully selected abstract and concrete nouns. Using Representational Similarity Analysis, we find that the implicit representational space of participants and the semantic representations of language models are significantly aligned. We also find that both representational spaces are implicitly aligned to an explicit representation of concreteness, which was obtained from our participants using an additional concreteness rating task. Importantly, using ablation experiments, we demonstrate that the human-to-model alignment is substantially driven by concreteness, but not by other important word characteristics established in psycholinguistics. These results indicate that humans and language models converge on the concreteness dimension, but not on other dimensions.
One of the fundamental tasks in natural language processing (NLP) is dependency parsing, which involves analyzing the grammatical structure of sentences by establishing relationships or dependencies between words. In this paper, we examine the difficulties and methods for performing dependency parsing for Bangla text, a language with complex morphology and distinctive syntactic properties. The article addresses the value of dependency parsing in capturing the linguistic subtleties of Bangla sentences and their applications in various NLP tasks. A method has been proposed to develop dependency parsing on Bangla text using a graph-based approach. A parsing tree is generated from a directed graph using Bangla input. The proposed system is achieved overall 68% accuracy which is evaluated using Bangla dependency corpus. This method enriches Bangla language linguistic resources and annotated corpora, facilitating the language’s global use. The resultant tree is also evaluated using evaluation metrics.
Recent experimental studies have examined GOODNESS IS BRIGHTNESS and a host of other primary metaphors. However, complex mappings such as INTELLIGENCE IS BRIGHTNESS have been largely ignored, nor has there been any attempt to distinguish their effects from those of primary metaphors such as GOODNESS IS BRIGHTNESS. The current study assesses both the nonprimary metaphoric mapping INTELLIGENCE IS BRIGHTNESS and the well-documented primary metaphor GOODNESS IS BRIGHTNESS in a visual priming task. The study finds that a bright background encourages photos of faces to be rated as both more intelligent and wellintentioned, though the background does not significantly affect either attribute alone. This suggests that two metaphors with the same source domain can reinforce each other. The study also underscores the difficulty in assessing a non-primary mapping in isolation from other factors.
This essay deals with two colour-related adjectives, badius and baietus, in the Medieval Latin documentary sources of Catalonia studied by the Glossarium Mediae Latinitatis Cataloniae (GMLC). The collection of documental testimonies has been conducted through the lexical database Corpus Documentale Latinum Cataloniae (CODOLCAT), a digital corpus of the Latin texts of this linguistic domain. These sources of notarial and juridical nature contain a considerable amount of colour adjectives, which usually serves to identify and differentiate lexical elements within the same referential class. One of the most attested colour adjectives is badius “bay, brown”, frequently documented with its variant baius and always referring to equines. There are also few occurrences of baietus “brownish”, a lexical innovation derived from badius. To refine their precise definitions, this study explores the forms and uses of both adjectives, taking into consideration the contexts in which they appear. Additionally, this study highlights the importance of the integration of digital tools into lexicographical research and emphasizes the need to incorporate insights from other linguistic domains to achieve a more comprehensive understanding of the words under analysis.
The article examines speech culture as a key component of language competence among higher education students. The author emphasizes that mastering the norms of the literary language, adhering to ethical and stylistic standards in communication, is an indicator not only of a person’s general education but also of their readiness for professional and social interaction. The main components of speech culture are analyzed, including accuracy, clarity, logic, appropriateness, purity, expressiveness, and aesthetic quality of speech. Particular attention is paid to common violations of linguistic norms observed in the student environment: the use of colloquial, slang, and foreign words without necessity, unjustified calques, bureaucratic expressions, as well as syntactic and orthoepic errors. The article outlines the main causes of linguistic carelessness, such as low reading culture, the influence of social media, and the decline in linguistic standards in everyday and educational communication. The author proposes a number of pedagogical and methodological strategies aimed at cultivating a high level of speech culture among students. These include the integration of communicative training into the educational process, regular involvement of students in stylistic text analysis, and the activation of creative language practices. Examples of typical speech situations are provided to demonstrate the contrast between cultured and uncultured language use, highlighting the importance of speaker selfreflection in improving overall language competence. The relevance of the study is due to the growing importance of speech culture in the modern educational environment, where effective communication is a key component of a specialist’s professional training.
In recent years, syntactic and semantic analysis tools have become increasingly important in various subfields of Natural Language Processing (NLP). These tools enable automatic parsing of large-scale sentences in language corpora, allowing researchers to uncover syntactic structures and statistical regularities of a given language. This study focuses on the development and evaluation of syntactic parsing models for the Uzbek language, employing two widely used approaches: constituency parsing and dependency parsing. For constituency parsing, a rule-based system was developed to identify noun and verb phrases along with their internal constituents. For dependency parsing, a set of hand-crafted linguistic rules was created and applied to syntactically analyze simple Uzbek sentences. As a result of this work, a dependency-based syntactic treebank for Uzbek-Named UzTreebank was constructed. The treebank includes 20,000 automatically parsed simple sentences, of which 10,000 were manually annotated. Additionally, 36 syntactic templates of simple sentences were identified, and 50 linguistic rules were formalized and integrated into the system. The suboptimal performance of the system at its current stage is primarily attributed to the absence of hybrid modeling approaches and the limited size of the training corpus. The paper presents an overview of the rule-based architecture, parsing results, and the current stage of syntactic resource development for the Uzbek language.
The present study extends recent work on Universal Dependencies annotations for second-language (L2) Korean by introducing a semi-automated framework that identifies morphosyntactic constructions from XPOS sequences and aligns those constructions with corresponding UPOS categories. We also broaden the existing L2-Korean corpus by annotating 2,998 new sentences from argumentative essays. To evaluate the impact of XPOS-UPOS alignments, we fine-tune L2-Korean morphosyntactic analysis models on datasets both with and without these alignments, using two NLP toolkits. Our results indicate that the aligned dataset not only improves consistency across annotation layers but also enhances morphosyntactic tagging and dependency-parsing accuracy, particularly in cases of limited annotated data.