Search Results (210)

Gender bias embedded in legal and professional texts perpetuates systemic inequality, yet research on bias identification and mitigation remains largely confined to English. Morphologically rich languages such as Greek, where grammatical gender pervades nouns, adjectives, pronouns, and participles, present unique challenges that existing approaches fail to address. This paper elaborates on a systematic methodology primarily focusing on identifying and mitigating gender bias in Greek-language job advertisements and legal documents. To accomplish that task, we define a taxonomy of nine gender bias rules tailored to the linguistic properties of Greek and construct domain-specific annotated datasets comprising 90 expert-curated few-shot examples across both textual domains. Using these resources, we employ XML-structured prompt engineering with in-context learning (ICL)and systematically compare three classes of models: (i) commercial large language models (LLMs), namely Claude Sonnet 4.5 and GPT-5.2, (ii) two open-weight small language models (SLMs), Mistral Small (24B) and Ministral (14B), and (iii) Llama Krikri (8B), a Greek-native language model built on Llama 3.1 and fine-tuned on high-quality Greek corpora. For each input text, the system identifies biased expressions, maps them to specific bias rules, provides explanations, and generates a fully corrected inclusive version. Our experiments reveal substantial performance disparities across model scales and linguistic specialization, with LLMs demonstrating superior contextual reasoning and SLMs exhibiting systematic over-correction and grammatical errors in Greek morphology. We further introduce a critical meta-rule addressing gender agreement with named entities to prevent spurious corrections in legal texts referencing identified individuals. The findings highlight the importance of model scale, language-specific adaptation, and carefully designed prompting strategies for bias mitigation in underrepresented languages. Full article

Causal inference in text data has been a demanding objective in the field of natural language processing, mainly due to the intrinsic ambiguity and context sensitivity inherent in data, inducing uncertainty. Diminishing this uncertainty is essential in identifying reliable causal connections and advancing predictive consistency. In this research, we introduce an uncertainty-aware ensemble architecture that combines multiple text embedding schemes with both linear and nonlinear classifiers to boost causal text detection. Both sparse and neural-level embeddings were employed, and then combined it with an ensemble weighting approach based on two uncertainty estimation techniques, namely entropy-based and KL divergence-based. Unlike conventional ensemble methods with uniform or fixed voting strategies, our approach assigns weights inversely proportional to classifier uncertainty, ensuring that confident models exert greater influence on the final decisions. Our results show that TF-IDF, through its effective word frequency weighting scheme, consistently outperforms other embedding techniques, achieving better performance across both linear and nonlinear classifiers on both datasets (News Corpus and CausalLM–Adjective group). The experimental results show that our uncertainty-aware ensemble approach enhances both calibration and confidence predictions. Entropy-based weighting improves confidence in the case of linear classifiers with accuracy, F1-score, entropy and prediction confidence values of 94.3%, 94.0%, 0.382 and 0.774, respectively, while in the case of nonlinear classifiers the KL divergence-based weighting acquires a better performance with an accuracy of 97.6%, F1-score of 97.2%, KL Mean value of around 0.055 and LogLoss of 0.221. Full article

Gradable adjectives (long, happy) differ from absolute adjectives (spotted) in that they are dependent on context and speaker/listener perspective for their interpretation. Such context sensitivity may present challenges for individuals with autism spectrum disorder (ASD); however, this has never been investigated for these linguistic elements. In the current study, we asked adolescents with ASD or typical development (TD), who were part of a larger longitudinal study in which autistic characteristics, nonverbal cognition (NVIQ), and standardized language were also assessed, to sort pictures whose properties were either gradable or absolute. Adolescents sorted pictures on two occasions. In the second sorting, we manipulated the context by adding images representing one end of the scale to induce a shift in interpretation. Contrary to prediction, both groups demonstrated sensitivity to the context-specific properties by shifting their cutoffs of what counted as ‘long’ or ‘happy’ when the array was changed. Whereas NVIQ correlated positively with physical property shifts for the TD group, language measures correlated negatively with emotion property shifts for the ASD group. Autistic characteristics were not related to shift patterns in either group. Adolescents with autism are clearly able to take context into account when interpreting gradable adjectives; however, those with better language seem more focused on maintaining their cutoffs more than shifting them. Full article

Background/Objectives: Dementia, characterised by cognitive decline, significantly impacts language abilities. While the risk of dementia increases with age, it often manifests years before clinical diagnosis. Identifying early warning signs is crucial for timely intervention. Previous research has demonstrated that changes in language, such as reduced vocabulary diversity and simpler sentence structures, may be observed in individuals with dementia. This study investigates the potential of linguistic analysis to detect early signs of cognitive decline by examining the writing of Sir Terry Pratchett, a renowned author diagnosed with Posterior Cortical Atrophy (PCA), typically a form of dementia caused by Alzheimer’s disease. Methods: This study analysed 33 Discworld novels by Terry Pratchett, comparing linguistic features before and after a potential turning point identified through analysis of adjective type-token ratios (TTR). Results: A significant decrease in lexical diversity (TTR) was observed for nouns and adjectives in later works. Total wordcount increased, while lexical diversity decreased, suggesting a shift towards simpler language. This shift coincided with a decrease in adjective TTR below a defined threshold, occurring approximately ten years before Pratchett’s formal diagnosis. Conclusions: These findings suggest that subtle changes in linguistic patterns, such as decreased lexical diversity, may precede clinical diagnosis of dementia by a considerable margin. This research highlights the potential of linguistic analysis as a valuable tool for early detection of cognitive decline. Further research is needed to validate these findings in larger cohorts and explore the specific linguistic markers associated with different types of dementia. Full article

The focus in this corpus-based study is on a set of Spanish constructions formed with the verb of visual perception, ver ‘to See’, and a predicate adjective or participle. In addition to a clearly recognizable transitive schema, the set includes various instances featuring a reflexive clitic pronoun coreferential with the subject, some of which have been argued to evidence the grammaticalization of lexical ver into a univerbated semicopular verb (pronominal verse), meaning little more than ‘be’ in some examples, and proximate to the intransitive sense of English look in other cases. We trace the evolution of these constructions in data spanning the history of the Spanish language, from its recorded beginnings to the present. We establish the need to distinguish two constructional sources of change, namely, an old middle-reflexive and a younger reflexive passive. We draw attention to the “renewal” of the Latin deponent videri ‘appear, look, seem’, which can be said to have taken place in Spanish as a product of the passive-derived process of grammaticalization undergone by ver. And throughout the paper we address problems of analyzability, attributable to the superficially identical strings of words that characterize the constructional patterns with a reflexive morpheme. Full article

Traditional degree semantics approaches have aimed to pin down the inherent class of adjectives. This paper presents a novel dynamic perspective, where the classification of an adjective is dynamic and syntactically dependent. Using measurement theory and fuzzy set analysis, the proposed framework defines dynamic patterns of adjective classes with a set of axioms and integrates these patterns with syntactic structures to explain the flexibility and constraints observed in adjectival expressions. Employing Mandarin data, the paper illustrates how different syntactic constructions select specific adjective classes, thereby affecting their distribution and interpretation. This approach not only accommodates cross-linguistic variations but also provides a more comprehensive understanding of the semantics of adjectives. Full article

There are limited empirically supported interventions that target three outcomes—behavior of children with ASD (instead of using adjectives such as “disruptive,” “interfering,” “problem,” or “challenging” behavior, we use “behavior” to avoid ableist language), parent stress, and parenting sense of competence. To help address this need, we tested a hybrid telehealth adaptation of COMPASS for Hope (C-HOPE), an 8-week parent-mediated program originally offered via face-to-face or synchronous telehealth delivery. The present study incorporated asynchronous group discussion board sessions hosted on a learning-management platform together with synchronous individual coaching sessions by telephone. Using a pre-post design, 10 caregivers completed the intervention. Effect sizes were calculated for three treatment outcomes of child behavior, parent stress, and parenting sense of competence. There was a statistically significant difference in the scores for child behavior, with a large effect size (d = 0.73) and a statistically significant difference in parent stress, with a medium effect size (d = 0.50). No difference was observed for parenting sense of competence. Treatment adherence and caregiver satisfaction for the intervention were acceptable. Findings support the feasibility and promise of combining asynchronous and synchronous telehealth elements to increase access to evidence-based parent-mediated interventions for ASD. Full article

While extensive research has focused on ChatGPT in recent years, very few studies have systematically quantified and compared linguistic features between human-written and artificial intelligence (AI)-generated language. This exploratory study aims to investigate how various linguistic components are represented in both types of texts, assessing AI’s ability to emulate human writing. Using human-authored essays as a benchmark, we prompted ChatGPT to generate essays of equivalent length. These texts were analyzed using Open Brain AI, an online computational tool, to extract measures of phonological, morphological, syntactic, and lexical constituents. Despite AI-generated texts appearing to mimic human speech, the results revealed significant differences across multiple linguistic features such as specific types of consonants, nouns, adjectives, pronouns, adjectival/prepositional modifiers, and use of difficult words, among others. These findings underscore the importance of integrating automated tools for efficient language assessment, reducing time and effort in data analysis. Moreover, they emphasize the necessity for enhanced training methodologies to improve AI’s engineering capacity for producing more human-like text. Full article

This paper addresses a well-known puzzle at the intersection of morphology and syntax: the categorical exclusion of adjectives modified by evaluative morphology from prenominal position in Italian. While Italian allows many adjectives to occur both pre- and postnominally, adjectives like piccolino, ‘little-dim’, are strictly postnominal (cane piccolino, lit. ‘dog little-dim’ vs. *piccolino cane, ‘little-dim dog’), a distribution not fully explained by their proposed predicative or intersective nature. Drawing on degree semantics and trope theory, we argue that this constraint arises from an incompatibility between two distinct interpretive strategies. Prenominal adjectives undergo a syntactically driven semantic shift, whereby the noun triggers a trope-based interpretation of the adjective, redefining the meaning of the A-N complex. In contrast, evaluative morphology operates through a pragmatically driven strategy, contributing speaker-oriented, context-sensitive meaning to the adjective. Crucially, these two strategies are mutually exclusive: an adjective modified by evaluative morphology has already undergone pragmatic reinterpretation and cannot simultaneously participate in the compositional syntactic process required for prenominal placement. This explains why adjectives with evaluative suffixes are excluded from prenominal contexts, despite often yielding intersective interpretations postnominally. Our proposal accounts for this distributional asymmetry without resorting to stipulations and suggests that certain interpretive procedures are not recursively applicable across syntax and pragmatics. Ultimately, this study sheds new light on a principled interface constraint linking syntactic distribution, morphological derivation and pragmatic interpretation. Full article

Excessive skin-picking has been conceptualized as a maladaptive emotion regulation strategy. One potential contributor to emotion regulation difficulties is low emotional granularity (EG), defined as the ability to precisely differentiate between emotional states. To our knowledge, this is the first study to investigate whether EG for unpleasant feelings is associated with the severity of skin-picking behavior. A total of 143 individuals (mean age = 25 years, 84% female) participated in an app-assisted one-week experience-sampling study. Twice daily, they provided adjectives to describe their current affective state (performance-based EG) and rated their urge to engage in skin-picking. Additionally, they completed a Skin-Picking Scale (SPS) and an EG questionnaire (self-reported EG). Results showed that higher SPS scores were associated with lower self-reported EG (B = −0.05). However, higher performance-based EG for unpleasant feelings was linked to higher SPS scores (B = 0.02), a greater urge to engage in skin-picking (B = 0.05), and a longer duration of the behavior (B = 0.01). The two EG measures were not correlated (r = 0.01). In conclusion, these findings suggest possible biases in self-perceptions of EG in those who excessively pick their skin. Interventions that train attentional focus and promote the valuing of affective diversity may help align self-reported and performance-based EG, and in turn reduce skin-picking. Full article

This research aims to design a mobile-based application for waste management in Sukabumi City using the Design Thinking method. This application is designed to help Sukabumi City residents find out waste transportation schedules, TPS locations, and information on waste volume. This research uses the Design Thinking stages which include Empathize, Define, Ideate, Prototype, and Test. The results of prototype testing using the System Usability Scale (SUS) showed an average score of 77.0580475, which was included in the “Acceptable” category with the adjective “Good”. This application is expected to increase public awareness of waste management and provide more regular information regarding waste transportation schedules. Suggestions for developing this application are to expand the scope of use of the system and pay attention to input from users to improve the quality of the application. Full article

With the advancement of haptic technology, the use of pseudo-haptics to provide tactile feedback without physical contact has garnered significant attention. This paper aimed to investigate whether sliding fingers over onomatopoetic text strings with pseudo-haptic effects induces change in perception toward their symbolic semantics. To address this, we conducted an experiment using finger-point reading as our subject matter. The experimental results confirmed that the “neba-neba,” “puru-puru,” and “fusa-fusa” effects create a pseudo-haptic feeling for the associated texts on the “hard–soft,” “slippery–sticky,” and “elastic–inelastic” adjective pairs. Specifically, for “hard–soft,” it was found that the proposed effects could consistently produce an impact. Full article

This eye-tracking study examined how heritage speakers of Spanish process gender agreement morphology at a distance, focusing on the activation of the gender feature during sentence processing. Previous work is conceptually replicated and further extended by assessing (1) whether reduced sensitivity to gender agreement mismatches when another word intervenes between the head noun and its modifying adjective stems from weakened gender feature activation, (2) whether a gender-marked determiner enhances this activation, and (3) whether Age of Onset of Bilingualism (AOB) plays a role in this activation. Fifty-three English-dominant heritage speakers of Spanish and a comparison group of 32 Spanish-dominant monolingually raised speakers read sentences with and without gender agreement mismatches while their eye movements were monitored. Sentences contained mismatches in adjectives modified by the intensifier “muy” under two conditions: a No Cue condition (e.g., árboles muy altos/*altas) and a Cue condition with a gender-marked determiner (e.g., unos árboles muy altos/*altas). Statistical modeling of the eye-tracking data suggests similar effects for both groups in the No Cue condition, but AOB and proficiency modulated sensitivity for heritage speakers with a later AOB (4–6). Gender cues on the determiner (Cue condition) impacted the time course of agreement processing for all groups, the total time spent reading mismatches for all heritage speakers as a function of proficiency, and the rereading time for heritage speakers with a later AOB (4–9). We consider the role of Age of Onset of Bilingualism (AOB) and proficiency in morphosyntactic processing, feature retrieval, and cue facilitation in heritage language processing. Full article

The Reformed Church in Hungary is a Reformed Protestant church in terms of its confession of faith, in which both adjectives, Reformed and Protestant, are emphasized. From this formulation follows the critique and firm rejection of a form of organization that existed before and still exists today: that of the Catholic Church. The foundations of Reformed (in this article, the term “Reformed” is used to designate the ecclesiastical and theological tradition associated with Calvin, Bullinger, Zwingli, and others) church institutions and church organization were formulated by Calvin in the Institutio, from which Reformed church law, through its historical development, formulated the principle of universal priesthood as a fundamental principle, the principle of synodal presbyterate as a constitutional principle of the church, and the principle of a free church in a free state, although the latter establishes the relationship between church and state. In distinguishing between a theologically postulated church and a church embodied in legal organization, canon law may examine the latter, and in particular, the canon law of the Protestant churches indeed sharply distinguishes it from the theological concept of church. Thus, in examining the development of the organization of the so-called visible church and the questions of the structure and functioning of the institution in the present, I will examine the organization and functioning of the Reformed Church in Hungary in the light of the organizational principles and methods that have developed historically, with a view to outlining the conditions for future optimal functioning. In my study, I trace the transformation of the Reformed Church from its beginnings to the change of regime. Full article