Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
16504 papers
In this paper, we as teachers and researchers present our analyses of pedagogically productive talk within a decolonial writing space.As mentors, we volunteered for a not-forprofit organization and worked with students on their writing, who come from an indigenous community in India.We analyze our bi-monthly mentor meetings, where we worked through course design, our challenges, and strategies.Our reflections were mediated by classroom videos that we watched together in these meetings.As part of our findings, we present two pedagogical deliberations that created the most tension from a decolonial lens.These deliberations included decisions around using the frame of counter stories, the scope of our intervention as mentors viz a viz linguistic norms and navigating the layers of distance between us and our students.Our findings have implications for training mentors and educators who are interested in designing decolonial learning environments.
This study investigates the emotional characteristics of the erhu and violin, focusing on the differences in emotional intensity. The experiment involved 25 participants categorizing the emotions and providing valence-arousal ratings for 14 musical excerpts. The results showed significant agreement between the two models, with the categorical emotions used to determine the emotional labels. A subsequent 9-point Likert comparison with 20 participants revealed that the instrument factor influenced emotional intensity. Overall, the violin conveyed more happiness, agitation, and calmness than the erhu, particularly for the Chinese-style excerpts. However, the sad emotional intensity was similar between the two instruments, contrary to the expectation that the erhu would convey sadder emotions. The performance factor had minimal impact on emotional intensity within the same instrument group, suggesting the inherent qualities of the erhu and violin were the primary drivers of the differences. Listeners more familiar with the instruments were likely to rate the violin higher in valence and arousal than the erhu. These findings contribute to understanding the expressive capabilities of the two instruments and highlight the instrument-specific emotional characteristics that shape the perceived emotional characteristics of music.
This paper presents the first attempt to automatically annotate Enhanced Universal Dependencies for Brazilian Portuguese. We use a symbolic annotation system, based on graph rewriting rules, and modify its original rules to better suit the linguistic characteristics of Portuguese using a manually annotated sample from the journalistic portion of Porttinari treebank as ground truth. Our objective is to assess the performance of the automatic annotation for a novel language and to determine the extent of possible improvements through rule modifications. Results demonstrate significant performance enhancements, where linguistic-driven rule adjustments improved the annotation accuracy 11.38 points, achieving 96.05% F1-score.
Discourse Analysis aims at high-level semantic and structural analysis. Discourse structure analysis and relation recognition are two key tasks in discourse analysis research, while discourse analysis plays an important role in studying the structure and semantic content of texts. This paper firstly introduces the Rhetorical Structure Theory (RST) and the Penn Discourse TreeBank (PDTB) annotation standards and its corresponding resources establishment of each corpus. Then the mainstream models of discourse structure and relation are expounded. In the last part, the opportunities and challenges of discourse analysis are discussed by combining with the mainstream large language model ChatGPT. In addition, the development prospect of discourse analysis is explored by summarizing the related studies.
Annotates Finnish textual survey responses into CoNLL-U format using Finnish treebanks from <<a href="https://universaldependencies.org/format.html" target="_top">https://universaldependencies.org/format.html</a>> using UDPipe as described in Straka and Straková (2017) <<a href="https://doi.org/10.18653%2Fv1%2FK17-3009" target="_top">doi:10.18653/v1/K17-3009</a>>. Formatted data is then analysed using single or comparison n-gram plots, wordclouds, summary tables and Concept Network plots. The Concept Network plots use the TextRank algorithm as outlined in Mihalcea, Rada & Tarau, Paul (2004) <<a href="https://aclanthology.org/W04-3252/" target="_top">https://aclanthology.org/W04-3252/</a>>.
Touch-mediated affect has largely been studied using natural textures and brushing techniques, which present challenges in control and deployment across haptic applications. A common alternative is vibrotactile stimulation (VBT) since it is easily deployable and accessible across devices. However, sparse literature exists on the VBT-induced affect modulation and its cortical correlates. Addressing this gap, we developed a novel paradigm that examined the behavioral and electrophysiological correlates of affect induced by VBT. We used electroencephalography (EEG) to record the cortical responses to VBT across six different locations and measured the concurrent affect ratings. Mixed effects modelling, an unsupervised modelling technique, was used to decipher the relationships between stimulation conditions, affect ratings and cortical modulations. Our study revealed that altering the duration of the vibrotactile stimuli can elicit distinct affect responses. The location of the VBT stimuli did not play a part in the perceptual aspects of affect but was involved in the cortical encoding of affect. Furthermore, early cortical processing in the somatosensory cortex (SCx) primarily encoded Arousal and not Valence. Our study lays out phenomenological aspects of cortical VBT processing and lays the groundwork for future research.
Social identities are created from the organisation of a series of coordinates that cross different areas in a community: the characteristics of the social group to which the individual belongs, the position that this individual has within the group, the social attitudes towards the own group and other groups, the type of activity that takes place (public or private), the linguistic policies existing in the community towards the different linguistic norms that coexist in it, etc. By incorporating all these (and other) aspects to the variationist analysis, we are admitting that neither the strictly structuralist nor the strictly interactional positions in Sociolinguistics allow us to properly explain the social dimension of language. The analysis of these relationships allows us to analyse with better criteria the different levels in which the sociocultural meaning that the forms of language acquire in specific social situations is organised. Within the framework of these ideas, this research analyses the way in which six radio broadcasters from the Canary Islands stylise their speech in order to achieve certain communicative purposes.
Languange holds a significant role in facilitating human thinking, serving as a foundation for understanding and accessing knowledge. However, the development of Indonesian language is currently experiencing a decline, mainly due to the widespread influence of social media. Social media users, often referred to as netizens, often use terms or vocabulary that are not in line with linguistic norms. This results in the communication patterns of Indonesian people in daily life, both in oral and written forms. This study uses an observation method with reading and note-taking techniques to describe the development of slang used by millennial teenagers on various social media platforms. The research findings show that the slang used by millennial teenagers comes from various sources, including regional languages, Indonesian, foreign languages, and a combination of Indonesian and foreign languages. Therefore, the use of slang by millennial teenagers is interpreted as a form of self-expression when building friendships and fostering close relationships between fellow teenagers
This article explores the dynamic evolution of language within Generation Z, focusing particularly on the impact of social media and digital communication on linguistic practices among Polish youth. The study examines the morphological aspects of Gen Z slang, including acronyms, clippings, borrowings, derivations, and compound formations. Through a comprehensive analysis of linguistic data collected from online repositories, the study uncovers recurring morphological processes and sheds light on the underlying motivations and cultural influences driving language change within this demographic. Findings reveal Polish Generation Z’s creativity, and global connectivity in shaping linguistic norms and expressions. The study emphasizes the dynamic nature of language evolution within digital contexts, highlighting the role of younger generations as agents of linguistic innovation. This study contributes to the understanding of contemporary language dynamics and offers valuable insights into the evolving landscape of human communication in the digital age, specifically within Polish context.
Comments on an article by Jay W. Schwartz, Kayleigh H. Pierson, and Alexander K. Reece (see record 2024-19488-001). In this issue, Schwartz et al. (2024) tackle the pitch rule in humans by testing to what extent we use pitch alone to judge emotional arousal across closely and distantly related animal species. The findings of Schwartz et al. open a number of intriguing possibilities for future research: Notably important additional steps would include to further investigate the accuracy of the pitch rule across closely and distantly related species. Upon this, in order to study the evolutionary ancestry of the pitch rule, it will be necessary to study its applicability across nonhumans. Particularly interesting would be the inclusion of subject species that have been found to eavesdrop on heterospecific alarm calls. Previous research (see Hoeschele, 2017 for a review) as well as present findings on human ratings of macaque versus cricket calls also suggest that we should additionally focus on sound features that compliment emotional arousal rating beyond pitch such as spectral information. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
Previously used to refer to generic antecedents and antecedents of unknown gender, singular they has been found to increasingly occur with definite antecedents of known gender. This shift is associated with rising awareness of nonbinary gender identities and the expansion of they as a preferred pronoun. Usage of singular they has been previously examined only within Inner Circle Englishes (e.g., US English). In this study, we investigate sociolinguistic factors that influence the acceptability of singular they in Singapore English, an Outer Circle variety that is pivoting towards internal linguistic norms but also experiences frequent contact with non-local Englishes. We find that singular they is rated as significantly more grammatical by younger respondents; its rating is also constrained by definiteness and interactions between social factors, including gender and religiosity. These factors are found to be stronger predictors of singular they acceptability than linguistic prescriptivism. The diffusion of singular they to Singapore English illustrates the ongoing role of non-local contact in the evolution of this variety.
ABSTRACT This study aims to examine the use of curse words in the Serawai language in Tebat Gunung Village, Semidang Alas District, Seluma Regency. The study focuses on the forms and meanings of curse terms used by the local community, thereby contributing to a broader understanding of language as a medium for expressing ideas and thoughts in accordance with linguistic norms. The method employed is qualitative research with a descriptive approach, aiming to deeply and comprehensively understand the studied phenomenon. The results show that curse words remain a significant part of daily communication among the Serawai community. These words are used in emotional contexts, such as insults or mockery, as well as in joking contexts. A total of 60 curse words were identified during the study in Tebat Gunung Village. In conclusion, curse words in the Serawai language reflect the unique cultural habits of the local community, which should be viewed in their social and cultural context. Keywords: Serawai Language, Tebat Gunung Village, Curse Words
This chapter describes the courageous pedagogical choices made by the author to design and implement a teaching module entitled, “African American English Ain't Broken: The Linguistic Dexterity of Black Folks” for pre-teacher education students enrolled in an multicultural education class in her university's School of Education. The author describes her intention to focus on Black English as a unique variety of English as an act of resistance to white, hegemonic linguistic norms in teacher education. Grounded in Critical Theory and designed using the principles of Culturally Responsive Teaching, this teaching module enables pre-education students to apply their new knowledge of the nature of Black English to develop an understanding of the systems of power and oppression at work in schools and society that marginalize speakers of African American English, including K-12 students. The chapter concludes with suggestions for ways teacher educators, and K-12 classroom teachers can check their linguistic biases and honor students' home languages in meaningful ways.
Translating between languages with drastically different grammatical conventions poses challenges, not just for human interpreters but also for machine translation systems. In this work, we specifically target the translation challenges posed by attributive nouns in Chinese, which frequently cause ambiguities in English translation. By manually inserting the omitted particle X ('DE'). In news article titles from the Penn Chinese Discourse Treebank, we developed a targeted dataset to fine-tune Hugging Face Chinese to English translation models, specifically improving how this critical function word is handled. This focused approach not only complements the broader strategies suggested by previous studies but also offers a practical enhancement by specifically addressing a common error type in Chinese-English translation.
The article focuses on the peculiarities of translating English abstract nouns Singularia Tantum used in literary texts into the Ukrainian language. In particular the paper aims to find out and explicate factors that determine the number form of a Ukrainian abstract lexeme in the case if it can obtain both number forms. Having analysed the semantic structure of polysemantic abstract nouns Singularia tantum, linguists noticed that some abstract lexemes developed new meanings in the plural, and the correlation between their singular and plural is not paradigmatic. That is why linguists refer to such plural form of an abstract noun as lexicalised plural. It was observed that nouns denoting negative feelings acquired additional semantic meanings of intensity, duration in the plural. Therefore, lexicalised plural of the mentioned abstract lexemes are often used by interpreters to create certain stylistic effects in the target text. The author highlights that nouns of low degree of abstraction are able to obtain both number forms due to the presence of generalised and specific meanings in their semantic structure. While generalised meaning denotes state, quality, or action, the specific one signifies a specific manifestation of state, quality, or action. It should be noted that it is the specific meaning that can be used in both number forms. While the singular of these nouns denotes one specific manifestation of state, quality, or action, the plural signifies several manifestations. Since the mentioned nouns do not acquire new meanings in the plural, it is possible to substitute generalised meanings with specific ones, and vice versa while translating. Thus interchangeability of number forms can be traced in the Ukrainian translation. The research emphasises that the possibility of the mentioned interchangeability is limited by linguistic norms of the target language and the referential status of a source abstract lexeme in an utterance. In particular, Ukrainian quantitative structures prefer the plural of a noun that is quantified. The linguistic norms of the Ukrainian language permit some abstract nouns to use their singular and plural synonymously. It also has been mentioned that grammatical number forms of a noun acquiring a referential status cannot be changed while translating since the number of manifestations of quality, state, or action is relevant to the speaker. On the other hand, a translator can use both number forms of a Ukrainian equivalent if a source abstract lexeme acquires a non-referential status.
The paper deals with pragmatic restrictions treated as limits of normal communication within certain situations. Violations of such limits result in communicative failure or switch the communication into a sphere of jokes. Such restrictions correlate with formal and semantic linguistic norms, they have a gradual nature and vary from absolute prohibitions to admissible, but not recommendable communicative actions. Pragmatic restrictions are ethnoculturally bound. They change in time, as new norms emerge and previous rules function no longer. Several types of pragmatic restrictions may be singled out in communication: 1) etiquette rules which fix the canons of demonstrating good will and respect; 2) fundamental norms of behavior which keep the most important communicative formats, mainly concerning religious prescriptions and prohibitions; 3) common sense rules in everyday personal communication; 4) ritual automatic adherence to traditional beliefs; 5) politically correct, which is a reaction to violations of the rights of an oppressed or vulnerable part of the population; 6) discourse and style norms characterizing situationally defined norms of communicative behavior.
This study characterizes linguistic sophistication within university-based online courses using Learning Management Systems (LMS) across various academic disciplines. The research employs natural language processing tools to extract detailed linguistic features from student discussion posts and utilizes Principal Components Analysis (PCA) to identify distinct linguistic profiles. These profiles are analyzed to understand how linguistic sophistication varies across different educational contexts, specifically among various schools and courses. Subsequent cluster analysis reveals statistically significant distinct groups based on linguistic attributes. Despite the comprehensive analysis, the study did not establish significant predictive models linking linguistic sophistication to any direct educational outcomes. Instead, the findings highlight significant differences in language use across disciplines, suggesting that each academic field may have unique linguistic norms. The study emphasizes the need for further research to explore the underlying factors that influence these linguistic characteristics and their implications for educational practices.
Explicitation constitutes a focal point within the realm of interpretation studies. It embodies a process whereby expressions originating from the source language are rephrased into clearer and more comprehensible renditions, harmonizing with the linguistic norms and practices of the target language. The process of explicitation involves not only the conversion of languages but also a deep understanding of cultural backgrounds and context. Explicitation goes beyond mere language conversion, effective explicitation enhances the precision and subtlety of meaning transfer in interpreting. Most of the research on explicitation in the academic circle focuses on translation. This paper endeavors to furnish a thorough examination of explicitation research through a multifaceted lens, encompassing the delineation of explicitation, its classification, as well as both domestic and international scholarship pertaining to explicitation, alongside an exploration of the underlying rationales. Its primary objective is to serve as a scholarly resource for forthcoming investigations into the phenomena of explicitation in interpretation, thus stimulating translators within our jurisdiction to assume a more pronounced role in this domain and actively participate in discourses on explicitation research within the global translation community.
This thesis explores the utilization of non-common expressions in English mass media, investigating the implications of this linguistic phenomenon on public discourse and perception. Drawing upon a range of contemporary media sources, from printed newspapers to digital platforms, the study analyzes how unconventional expressions, including idiomatic phrases, industry jargon, and neologisms, shape narratives and potentially influence readers’ understanding of events and issues. The research engages with theories of communication and linguistics to examine the functions and effects of these expressions, looking at their frequency, contexts, and the dynamics between traditional linguistic norms and the evolving language within the public sphere. Clear evidence is provided that the use of non-common expressions not only captures attention and conveys complex ideas succinctly but also serves to frame issues in specific ways, carrying subtle connotations and cultural references. This study thereby sheds light on the role of language innovation in journalism and its broader social implications, offering insights into the strategic use of language in mass media.
This paper examines the attitudes of Montenegrin students toward dialectal language forms in communication. An empirical study was conducted for this purpose, using a questionnaire with eight questions. The interpretation of the results provided important insights into the factors shaping the attitudes of Montenegrin students on this issue. The responses revealed value judgments indicating that, for a high percentage of the student population, it is important to prioritize forms from the standard language in almost all types of communication. The study found that most of the responses reflect attitudes shaped by a traditionalist approach to linguistic norms, where the linguistic standard is idealized as a model of cultural prestige. Considering the findings of this research, it is concluded that a potential measure to address this situation would involve systematically designed education programs developed through broad collaboration between institutions and non-governmental organizations, targeting young people and society as a whole. An important contribution to dispelling linguistic myths and unscientific attitudes regarding non-standard dialects is seen in the legal protection of Montenegrin language varieties as intangible cultural heritage.
The article reflects on the historical and socio-linguistic processes in Ukraine that contributed to the liberalization of language norms in the late 20th and early 21st centuries. This has led to the emergence of neologisms. There are factors that were shaping the linguo-cultural space in Ukraine during this period: granting Ukrainian the status of the sole state language, as the language of the titular nation of Ukraine, legislative and linguistic initiatives that supported the development of the language, globalization processes, and geopolitical changes. Abbreviative derivatives as neologisms of modern Ukrainian emerged during the process of liberalization and belong to four distinct parts of speech. Actually, speakers form adjectival derivatives. They are most frequently found in journalistic texts, as well as in conversational, literary, and academic styles. A socio-linguistic survey conducted in 2024 confirmed trends in the word formation of adjectival derivatives from abbreviative bases. Speakers add the confix (interfix + suffix) -ivs`k-. Today, the main issue in researching such neologisms is the identification, justification, and systematization of variant and invariant standard forms according to linguistic norms.
This study is part of the bigger project OPTIMAX - Optimizing Outcomes in Psychotherapy for Anxiety Disorders (Müller-Bardorff et al., 2022, https://psyarxiv.com/yezaj/). Self-efficacy is a key construct in behavioral science impacting mental health and psychopathology. Here, we expand on previously demonstrated between-person self-efficacy effects. We prompted 66 patients five times daily for 14 days prior to starting cognitive behavioral therapy (CBT) to provide avoidance, hope, and perceived psychophysiological arousal ratings. Multilevel logistic regression analyses confirmed self-efficacy’s significant effects on avoidance in daily life (OR = 0.53, 95% CI: 0.34–0.84, p =.008) and interaction effects with anxiety in predicting perceived psychophysiological arousal (OR = 0.79, 95% CI: 0.62–1.00, p =.046) and hope (OR = 1.21, 95% CI: 1.03–1.42, p =.02). More self-efficacious patients also reported greater anxiety symptom reduction early in treatment. Our findings assign a key role to self-efficacy for daily anxiety symptom experiences and for early CBT success. Self-efficacy interventions delivered in patients’ daily lives could help improve treatment outcome.
In the modern period social media has tremendously affected linguistic norms with the prevalent usage of Internet language, which has become an integral part of our digital interactions. From its modest beginnings in early chat rooms to its current ubiquitous presence across social media platforms, comprehending a new Internet term and delighting in its usage can open up an entirely new world of connections and expressions. This article explores the impact of internet language on daily communication. It analyses the main types of internet slang, highlighting the emergence of new linguistics strategies, including symbol graphics, emojis, numerical codes and abbreviated words. It also discusses the characteristics of internet slang and its impact on communication, emphasizing the numerous merits of internet languages, while acknowledging the challenges posed by their informal nature, such as the barriers to information dissemination. Additionally, the article underscores the need to adopt measures to improve the effectiveness of internet terminology in communication and manage its impact consciously and positively.
This study explores the learnability of memory-less and memory-augmented RNNs, which are theoretically equivalent to Pushdown Automata. Empirical results show that these models often fail to generalize on longer sequences, relying more on precision than mastering symbolic grammar. Experiments on fully trained and component-frozen models reveal that freezing the memory component significantly improves performance, achieving state-of-the-art results on the Penn Treebank dataset (test perplexity reduced from 123.5 to 120.5). Models with frozen memory retained up to 90% of initial performance on longer sequences, compared to a 60% drop in standard models. Theoretical analysis suggests that freezing memory stabilizes temporal dependencies, leading to robust convergence. These findings stress the need for stable memory designs and long-sequence evaluations to understand RNNs true learnability limits.
FrameNet serves as a comprehensive lexical database intended to represent contemporary language usage. However, it faces challenges in accurately representing specialized domains. Among these domains, FrameNet presents difficulties in capturing the specific semantics of human senses. Senses such as smell and taste are in fact included in more general frames or inadequately represented. Building on a previous resource proposing a new framework for olfactory events, we propose a similar annotation scheme for gustatory references in English, enlightening the potential of frames to effectively capture sensory semantics. Having a comprehensive framework to deal with the annotation of this kind of references in textual data is especially important to develop systems for the automatic extraction of sensory information. Moreover, our approach incorporates words from specific historical periods, thereby enriching the framework’s utility for studying language in a diachronic perspective. In this...
This paper addressed the practical application and potential impact of WordNet as an educational tool for teaching the lexical semantics of English nouns. WordNet, an extensive English lexical database, systematically organizes words into semantically interconnected groups and hierarchical classifications, providing a promising resource for utilizing a computational semantics approach in language education. Using the descriptive research method and the comparative method, the study explored the network of semantic relationships within English nouns, specifically focusing on synonymy, polysemy, hyponymy, and hypernymy. Based on the WordNet description, the paper suggested some practical techniques for teaching and learning lexical semantics within English nouns. These approaches aim to enhance the educational experience for both teachers and students by fostering a deeper understanding of the semantic relationships of English nouns through addressing readability, contextual meanings, and conceptual structures.
This study examined the extent to which musical training, familiarity, and personality predict music preferences among Malaysian secondary school students. Subjects were 381 16-year-old Malaysian secondary school students from Pasir Gudang, Johor divided into two groups, i.e. those with musical training and those without musical training. The subjects listened to forty music excerpts from eight genres, including Malaysian Pop, Western Pop, Malaysian Hip-Hop, Western Hip-Hop, Malaysian R&B, Western R&B, Malaysian Rock, and Western Rock. Only vocal excerpts were utilised, with tempos ranging from 140 to 180 beats per minute. Subjects completed three questionnaire sections. Section A was demographic information; Section B consisted of the music preference inventory (MPI) and familiarity rating scale; and Section C was the Big Five Inventory (BFI). Results showed that musical training had a strong influence on music preference. Subjects with musical training obtained higher mean scores for each music genre preference. Familiarity was also strongly correlated with music preferences. Each of the Big Five personality dimensions was also positively correlated with every music genre preference. In conclusion, musical training, familiarity, and personality play a crucial role in music preference decisions.
Temporal Convolutional Networks (TCNs) are one-dimensional convolutional neural networks for modelling sequential data. A key component in TCN is the dilation, that is used to increase the receptive field while keeping the number of parameters low. Dilation rates are predetermined in TCN. In this paper, an adaptive method is introduced for learning dilation rates by utilizing trainable binary masks with sparsity constraints (named as Adaptive TCN, AdaTCN). To select connections that are deemed important, the binary masks are applied to convolutional layers. We introduce structured sparsity into the mask using Gumbel Sharp softmax in order to control the number of active connections. Four different models, including TCN, random masked TCN, AdaTCN, and an AdaTCN that is initiated with TCN-like mask are trained and evaluated. With the Penn TreeBank (PTB) and WikitText-2 (WT2) datasets, experiments are conducted on word-level language models
Abstract The wider availability of large-scale datasets and reproducible algorithms has boosted the application of NLP to living languages. On the other hand, dead languages benefit from the availability of curated resources both to offset the sparseness of available data and to make data accessible to researchers. We present here AGVaLex, a computational valency lexicon automatically extracted from the Ancient Greek Dependency Treebank. It contains quantitative corpus-driven morphological, syntactic and lexical information about verbs and their direct and indirect arguments and has a wide range of applications for the study of Ancient Greek. To illustrate these applications, we offer a case study that compares the semantic flexibility of transitive verb formulae in archaic Greek epic to a non-formulaic corpus, with the goal of detecting unique patterns of variation. We also illustrate the possibilities afforded by AGVaLex to scholars with a less extensive background in computational corpus-based research.
Revealing the syntactic structure of sentences in Chinese poses significant challenges for word-level parsers due to the absence of clear word boundaries. To facilitate a transition from word-level to character-level Chinese dependency parsing, this paper proposes modeling latent internal structures within words. In this way, each word-level dependency tree is interpreted as a forest of character-level trees. A constrained Eisner algorithm is implemented to ensure the compatibility of character-level trees, guaranteeing a single root for intra-word structures and establishing inter-word dependencies between these roots. Experiments on Chinese treebanks demonstrate the superiority of our method over both the pipeline framework and previous joint models. A detailed analysis reveals that a coarse-to-fine parsing strategy empowers the model to predict more linguistically plausible intra-word structures.
The paper reports on a series of experiments aiming at probing LeBenchmark, a pretrained acoustic model trained on 7k hours of spoken French, for syntactic information. Pretrained acoustic models are increasingly used for downstream speech tasks such as automatic speech recognition, speech translation, spoken language understanding or speech parsing. They are trained on very low level information (the raw speech signal), and do not have explicit lexical knowledge. Despite that, they obtained reasonable results on tasks that requires higher level linguistic knowledge. As a result, an emerging question is whether these models encode syntactic information. We probe each representation layer of LeBenchmark for syntax, using the Orféo treebank, and observe that it has learnt some syntactic information. Our results show that syntactic information is more easily extractable from the middle layers of the network, after which a very sharp decrease is observed.
Public discourse often excludes the erroneous speech of migratory subjects, thus foreclosing social and political rapprochement. This paper argues that the position outside of pregiven, linguistic norms provides migratory speech with an improvisatory quality that can serve as a catalyst for community formation. Simulating freestyle forms in writing, Feridun Zaimoğlu’s Kanak Sprak seeks to find a new language for the critique of xenophobia and to establish belonging based on precarious conditions. In a close reading of Fikret’s monologue “Pity is that true vitamin,” I show how improvisation disrupts established discourses and transforms the meaning of conventional hate speech tropes to forge transethnic alliances. The paper then turns to the subsequent volume Koppstoff and problematizes the commodification of Kanak speech in neoliberal pop culture. Çağıl’s monologue “If you’re smart, you take our side” hints at a different understanding of improvisation that reframes the relation between mainstream society and its others. Drawing on critical improvisation studies, the paper contributes to the understanding of linguistic interventions into social orders that determine who can say what, in which speech form, and according to which norms of belonging.
The article discusses major characteristics of Norwid’s language and style in light of the concept of “hyper-grammaticality” [po-nad-gramatyczność] developed by the poet himself. Considered as a descriptive category, it organizes and foregrounds certain properties of his syntax as well as other elements. The adjective “hyper-grammatical” can be understood in three ways: 1. failing to comply with rules; 2. departing from linguistic convention; hence unconventional; 3. derived from a different level of language than grammar.Norwid’s works can be shown to display hyper-grammaticality in all of the above senses. Discussion of constructions that violate linguistic norms accounts for the following: anacoluthon,homonymic structures, obscurities related to functions of anaphoric elements, and disruptions of coherence. Unconventional elements departing from the epoch’s standards include, among other things, innovations in collocability, complications of syntax, numerous parenthetical remarks, and the usage of archaic constructions. In Norwid’s texts an important place is held not only by mechanisms proper to syntactical or grammatical level of enunciation, but also by phenomena present on other levels:meta-textualityandthematic-rhematic structure.
Public discourse often excludes the erroneous speech of migratory subjects, thus foreclosing social and political rapprochement. This paper argues that the position outside of pregiven, linguistic norms provides migratory speech with an improvisatory quality that can serve as a catalyst for community formation. Simulating freestyle forms in writing, Feridun Zaimoğlu’s Kanak Sprak seeks to find a new language for the critique of xenophobia and to establish belonging based on precarious conditions. In a close reading of Fikret’s monologue “Pity is that true vitamin,” I show how improvisation disrupts established discourses and transforms the meaning of conventional hate speech tropes to forge transethnic alliances. The paper then turns to the subsequent volume Koppstoff and problematizes the commodification of Kanak speech in neoliberal pop culture. Çağıl’s monologue “If you’re smart, you take our side” hints at a different understanding of improvisation that reframes the relation between mainstream society and its others. Drawing on critical improvisation studies, the paper contributes to the understanding of linguistic interventions into social orders that determine who can say what, in which speech form, and according to which norms of belonging.
Deep neural networks (DNNs) are widely used in fields like computer vision and natural language processing. A key component of DNN training is the optimizer. SGD-Momentum is popular in many DNN methodologies, such as ResNet and DenseNet, due to its simplicity and effectiveness. However, its slow convergence rate limits its use. To overcome this, we introduce inter-gradient collision into SGD-Momentum, inspired by the elastic collision model in physics. This new method, called ICSGD-Momentum, aims to improve convergence. We provide theoretical proof of convergence and establish a regret bound for ICSGD-Momentum. Experiments on benchmarks including function optimization, CIFAR-100, ImageNet, Penn Treebank, COCO, and YCB-Video show that ICSGD-Momentum accelerates training and enhances the generalization performance of DNNs compared to optimizers like SGD-Momentum, Adam, Radam, Adabound, and AdaBelief.
Western tonal music uses harmonic cadences as structural and syntactic markers. This study examines cadence perception regarding the role of pre-cadential and cadential mode congruence. Major-to-minor and minor-to-major mode shifts that occur in Parallel-Mode Tonic (PMT) and Deceptive cadences were of interest, with the former including a key change. 61 adults with varied musical training rated arousal, valence, and degree of completion (DOC) for four cadence types (Authentic, PMT, Deceptive, and Neapolitan) in two pre-cadential modes (major or minor). Arousal ratings did not vary much across all cadence types and mode variations. In contrast, valence perception was predominantly influenced by the pre-cadential mode with major mode receiving higher ratings, and DOC perception further correlated with the final chord’s within-context tonal stability. Importantly, the perception of PMT cadences was unique in that the mode of the final tonic chord became far more important in both valence and DOC ratings, pointing to privileged and distinct perceptual processing when mode shifts resolve on a tonic chord, especially of major mode (i.e., Picardy Third). Musical training manifested as significantly enhanced rating tendencies which were most apparent for DOC. The results substantially extend our understanding of the interactive nature of modes, tonality, and expertise in music perception.
Abstract Understanding the somatovisceral responses to auditory affective imagery has important implications in disorders like misophonia. The current study compared physiological responses to aversive and nonaversive states across three modalities: audiovisual, auditory, and auditory imagery. Electromyographic activity over corrugator supercilii (EMGc) and zygomaticus major (EMGz), electrodermal activity (EDA), heart rate (HR), and finger skin temperature (SKT) were measured. There was significant differentiation in EMGc, EDA, and HR deceleration between aversive and nonaversive audiovisual stimuli. EMGc potentiation was the only physiological measure showing consistent differentiation across the three modalities. Cross-modal aversiveness classification results revealed a similar physiological response pattern between audiovisual and auditory modalities. The physiological response pattern during auditory affective imagery was useful for predicting the aversiveness in audiovisual modality but not the other way around. Vividness in auditory imagery correlated with subjective hedonic valence ratings, but not physiological responses. Taken together, the current data suggest that the aversiveness of auditory imagery is differentiable in subjective affective experience and facial muscle potentiation. These results of the physiological responses to imagined aversive sounds in nonclinical population would serve as a comparison baseline for the study of misophonia.
The only (non-divisive) way to address the representation of linguistic gender identity is to attempt – albeit not without difficulty – a negotiation between the protection of the common linguistic norm and accommodating the progressive proliferation of identities. This process must begin from a specific point, one for which no general consensus yet exists. There is no grammatical sacrilege in feminizing terms such as sindaco or ministro: in Italian, masculine nouns ending in -o typically take -a in their feminine forms, leaving no structural reason to reject sindaca or ministra. However, inclusive forms such as direttorə and pittorə, autorə and lettorə, rather than challenging their masculine equivalents, risk stalling progress. In classical Latin, the feminine form of pictor (pictrix) did not exist; a woman engaged in painting in ancient Rome had to resort to circumlocutions like pingendi artifex («an artist in the field of painting»). It took centuries to “institutionalize” many feminine forms of professional titles, and we are still far from granting them full social recognition. The current wave of advocacy for a neutral gender risks relegating to obscurity precisely those feminine forms (such as direttrice and pittrice, autrice and lettrice) that we must instead promote – and encourage others to promote – without hesitation.
This research discusses the shift of the Tae' language through a case study of language attitudes and language usage in the city of Palopo, South Sulawesi. The Tae' language, as the identity of the Luwu people, in this case, the city of Palopo, is rarely used as the daily communication language of the people of Palopo. The use of the language in the family environment, which should be the closest domain to the regional language, has also been replaced. The method used in this research is a descriptive qualitative method. This study uses the instrument Cohn et al. (2013) in the form of scoring and social factors that can contribute to attitudes. Data analysis of language attitude employing the concepts of Garvin and Mathiot which encompass characteristics such as language loyalty, language pride, and awareness of linguistic norms. Based on the analysis of the data, the researcher found that the Tae' language as the regional language of the people of Palopo or Luwu has experienced a shift. The people in the city of Palopo have a positive attitude towards the Tae' language, but its usage is still minimal, even within the family domain. The Indonesian language dominates the language usage among the people. The positive language attitude does not align with the positive language usage, and it can even be negative in its usage.
The role of the amygdala in unconscious emotional processing remains a topic of debate. Past lesion studies have indicated that amygdala damage leads to impaired electrodermal activity in response to subliminally presented emotional stimuli. However, electrodermal activity can reflect both emotional and nonemotional processes. To provide behavioral evidence highlighting the critical role of the amygdala in unconscious emotional processing, we examined patients (n = 16) who had undergone unilateral resection of medial temporal lobe structures, including the amygdala. We utilized the subliminal affective priming paradigm in conjunction with unilateral visual presentation. Fearful or happy dynamic facial expressions were presented in unilateral visual fields for 30 ms, serving as negative or positive primes. Subsequently, neutral target faces were displayed, and participants were tasked with rating the valence of these targets. Positive primes, compared to negative ones, enhanced valence ratings of the target to a greater extent when they stimulated the intact hemisphere (i.e., were presented in the contralateral visual field of the intact hemisphere) than when they stimulated the resected hemisphere (i.e., were presented in the contralateral visual field of the resected hemisphere). These results suggest that the amygdala is causally involved in unconscious emotional processing.
This chapter is part of a broader research project that investigates the functions of codeswitching in the online community of Russian-speaking immigrants living in Italy. The focus of the current contribution is playful manipulations of code hereby analyzed under the framework of digital code play. This play includes various semiotic resources, such as grammatical and semantic variations deployed by the communicants on the Internet and the metalinguistic comments surrounding them. Digital code play often occurs in contexts where social and linguistic norms are negotiated. The material analyzed here comprises comments from forums and Facebook groups dedicated to the communication of Russian speakers. The chapter targets three discussion topics on the forum, Russianitaly, and one post on the Facebook group, Russkie v Italii ʻRussians in Italyʼ. These topics and posts focus on language elements and language use and attracted the attention of many users who reported unusual mixing of languages in a playful way. A qualitative analysis of the forms of the play and the surrounding comments shows that users position themselves as multicompetent speakers: creative individuals empowered by multilingual and sociocultural knowledge who are able to combine and recontextualise (linguistic) resources from different sociocultural spaces to create new meanings and values.
In the growing domain of natural language processing, low-resourced languages like Northern Kurdish remain largely unexplored due to the lack of resources needed to be part of this growth. In particular, the tasks of part-of-speech tagging and tokenization for Northern Kurdish are still insufficiently addressed. In this study, we aim to bridge this gap by evaluating a range of statistical, neural, and fine-tuned-based models specifically tailored for Northern Kurdish. Leveraging limited but valuable datasets, including the Universal Dependency Kurmanji treebank and a novel manually annotated and tokenized gold-standard dataset consisting of 136 sentences (2, 937 tokens). We evaluate several POS tagging models and report that the fine-tuned transformer-based model outperforms others, achieving an accuracy of 0.87 and a macro-averaged F1 score of 0.77. Data and models are publicly available under an open license at https://github.com/peshmerge/northern-kurdish-pos-tagging.
In this study, we investigated whether pictures of natural hazards (i.e., climate change consequences) elicit automatic (negative) affective responses using a picture-word interference task. In picture-word interference tasks, affective pictures and words are paired such that picture valence and word valence match (congruent) or mismatch (incongruent). Participants classify the valence of words (or pictures; targets) via key presses. Corresponding congruency effects in response times or error rates thus indicate that pictures (or words; distractors) interfered with target processing, that is, distractors elicited an automatic affective response. Here, we assessed how 12 natural hazard (negative) and 12 intact landscape (positive) pictures (distractors) interfered with the classification of four affective words (targets) as negative or positive. The obtained congruency effects demonstrate that natural hazard pictures (showing landslides, hail, wildfires, or droughts) elicit automatic affective responses, even though their valence can only be inferred based on scene semantics. Further, this implicit affective response measure did not correlate with self-report valence or arousal ratings for corresponding affective pictures, suggesting differences in the affective processes underlying implicit and explicit affective response measures. We conclude that picture-word interference tasks are a suitable means for determining implicit affective responses to even complex pictures, here negative affective responses to natural hazard scenes. This method thus also lends itself to investigations of affective responses in the context of climate change.
When encountering a potential threat, humans and animals engage in different strategic behaviors, such as orienting and defense, depending on the perceived threat imminence. Orienting has been associated with attentional immobility and heightened 'stimulus intake,' while defense is linked to action preparation and 'sensory rejection'. First, we replicated previous findings showing that humans exhibit either heart rate (HR) acceleration or deceleration in response to the same threat-related picture content. Second, we provide direct evidence that orienting, as indexed by increased HR deceleration, leads to enhanced visuocortical processing of threat-related images, as measured by steady-state visual evoked potentials (ssVEPs). Excitation of motor-relevant cortical circuits, assessed by beta-band desynchronization, was reduced in relation to HR deceleration. Conversely, HR acceleration was associated with a reversed pattern: reduced visual processing and increased excitation of cortical motor circuits, as reflected in ssVEP and beta-band modulations. While self-reported measures of state and trait anxiety, along with valence, arousal, and dominance ratings, did not account for variations in HR response patterns, faster avoidance motor responses were linked to defensive HR changes, and longer avoidance latencies were associated with orienting-like HR changes.
Second-order Recurrent Neural Networks (2RNNs) extend RNNs by leveraging second-order interactions for sequence modelling. These models are provably more expressive than their first-order counterparts and have connections to well-studied models from formal language theory. However, their large parameter tensor makes computations intractable. To circumvent this issue, one approach known as MIRNN consists in limiting the type of interactions used by the model. Another is to leverage tensor decomposition to diminish the parameter count. In this work, we study the model resulting from parameterizing 2RNNs using the CP decomposition, which we call CPRNN. Intuitively, the rank of the decomposition should reduce expressivity. We analyze how rank and hidden size affect model capacity and show the relationships between RNNs, 2RNNs, MIRNNs, and CPRNNs based on these parameters. We support these results empirically with experiments on the Penn Treebank dataset which demonstrate that, with a fixed parameter budget, CPRNNs outperforms RNNs, 2RNNs, and MIRNNs with the right choice of rank and hidden size.
Detection of out-of-distribution (OOD) samples is crucial for safe real-world deployment of machine learning models. Recent advances in vision language foundation models have made them capable of detecting OOD samples without requiring in-distribution (ID) images. However, 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 incorporates 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 extensive ablation studies and empirical evaluations, demonstrating state of the art performance of CLIPScope across various OOD detection benchmarks.
Verbs form the backbone of language, providing the structure and meaning to sentences. Yet, their intricate semantic nuances pose a longstanding challenge. Understanding verb relations through the concept of lexical entailment is crucial for comprehending sentence meanings and grasping verb dynamics. This work investigates the capabilities of eight Large Language Models in recognizing lexical entailment relations among verbs through differently devised prompting strategies and zero-/few-shot settings over verb pairs from two lexical databases, namely WordNet and HyperLex. Our findings unveil that the models can tackle the lexical entailment recognition task with moderately good performance, although at varying degree of effectiveness and under different conditions. Also, utilizing few-shot prompting can enhance the models' performance. However, perfectly solving the task arises as an unmet challenge for all examined LLMs, which raises an emergence for further research developments on this topic.