Search

There has been a growing effort to replace manual extraction of data from research papers with automated data extraction based on natural language processing, language models, and recently, large language models (LLMs). Although these methods enable efficient extraction of data from large sets of research papers, they require a significant amount of up-front effort, expertise, and coding. In this work, we propose the ChatExtract method that can fully automate very accurate data extraction with minimal initial effort and background, using an advanced conversational LLM. ChatExtract consists of a set of engineered prompts applied to a conversational LLM that both identify sentences with data, extract that data, and assure the data's correctness through a series of follow-up questions. These follow-up questions largely overcome known issues with LLMs providing factually inaccurate responses. ChatExtract can be applied with any conversational LLMs and yields very high quality data extraction. In tests on materials data, we find precision and recall both close to 90% from the best conversational LLMs, like GPT-4. We demonstrate that the exceptional performance is enabled by the information retention in a conversational model combined with purposeful redundancy and introducing uncertainty through follow-up prompts. These results suggest that approaches similar to ChatExtract, due to their simplicity, transferability, and accuracy are likely to become powerful tools for data extraction in the near future. Finally, databases for critical cooling rates of metallic glasses and yield strengths of high entropy alloys are developed using ChatExtract. Efficient data extraction from research papers accelerates science and engineering. Here, the authors develop an automated approach which uses conversational large language models to achieve high precision and recall in extracting materials data. [ABSTRACT FROM AUTHOR]

The task of cross-target stance detection faces significant challenges due to the lack of additional background information in emerging knowledge domains and the colloquial nature of language patterns. Traditional stance detection methods often struggle with understanding limited context and have insufficient generalization across diverse sentiments and semantic structures. This paper focuses on effectively mining and utilizing sentiment-semantics knowledge for stance knowledge transfer and proposes a sentiment-aware hierarchical attention network (SentiHAN) for cross-target stance detection. SentiHAN introduces an improved hierarchical attention network designed to maximize the use of high-level representations of targets and texts at various fine-grain levels. This model integrates phrase-level combinatorial sentiment knowledge to effectively bridge the knowledge gap between known and unknown targets. By doing so, it enables a comprehensive understanding of stance representations for unknown targets across different sentiments and semantic structures. The model’s ability to leverage sentiment-semantics knowledge enhances its performance in detecting stances that may not be directly observable from the immediate context. Extensive experimental results indicate that SentiHAN significantly outperforms existing benchmark methods in terms of both accuracy and robustness. Moreover, the paper employs ablation studies and visualization techniques to explore the intricate relationship between sentiment and stance. These analyses further confirm the effectiveness of sentence-level combinatorial sentiment knowledge in improving stance detection capabilities. [ABSTRACT FROM AUTHOR]

Recently, the issue of language in psychiatry has been raised more frequently. The concern for choosing the appropriate words has resulted in developing language recommendations regarding the way we should speak about treating mental health disorders. The lexeme psychiatryk 'mental hospital' has been mentioned among words that contribute to preserving the negative stereotypes of psychiatric care. Thus, the paper focuses on the ways in which this word is evaluated across the Internet. Since the author analyses specific utterances - the ways language is used, she employs the framework of linguistic pragmatics to account for the broad context of each utterance (the speaker's intentions, utterance type, the assumed hearer, etc.). In the majority of the analysed cases, we can observe such uses of the words that uphold the negative stereotype of institutionalized psychiatric care, and the hospitals themselves (in examples that exhibit playfulness, bordering on colloquialism and derision) are represented as appalling, pathological or funny and incomprehensible. [ABSTRACT FROM AUTHOR]

This paper contextualizes Okudzhava's song "And We to the Doorman" (AWD), initially marginal in the Soviet poetic mainstream. It explores its shifts in tone, irregular rhythms, colloquial language, and semi-criminal undertones. AWD's structure, with uneven stanzas and no clear refrain, reveals underlying symmetry and recurring themes. The meter is predominantly iambic but varies. Unconventional verse endings and various rhyme schemes, including distant chains, characterize its prosody. The narrative touches on social cohesion and class conflict. The style reflects a challenging attitude toward privilege, employing rhetorical devices and indirect threats. The melody aligns with thematic elements, featuring repetitive patterns and a spoken quality. Semantically, AWD presents an ambiguous message on class struggle and moral issues. In sum, this analysis uncovers Okudzhava's song's formal complexities, thematic nuances, and stylistic innovations. [ABSTRACT FROM AUTHOR]

Chat generative pretrained transformer (ChatGPT) is a conversational language model powered by artificial intelligence (AI). It is a sophisticated language model that employs deep learning methods to generate human-like text outputs to inputs in the natural language. This narrative review aims to shed light on ethical concerns about using AI models like ChatGPT in writing assistance in the health care and medical domains. Currently, all the AI models like ChatGPT are in the infancy stage; there is a risk of inaccuracy of the generated content, lack of contextual understanding, dynamic knowledge gaps, limited discernment, lack of responsibility and accountability, issues of privacy, data security, transparency, and bias, lack of nuance, and originality. Other issues such as authorship, unintentional plagiarism, falsified and fabricated content, and the threat of being red-flagged as AI-generated content highlight the need for regulatory compliance, transparency, and disclosure. If the legitimate issues are proactively considered and addressed, the potential applications of AI models as writing assistance could be rewarding. [ABSTRACT FROM AUTHOR]

It is a widespread view that 'academic language' is the most significant dimension of language. This aspect of language is widely used for academic purposes that contrast with the conversational language spoken outside of the classrooms. The ultimate focus on the linguistic features of academic syllabi overlooks the role everyday forms of conversational language play in academic work and how learners of marginalized backgrounds are capable of using their existing linguistic resources to navigate tasks within the classrooms. The paper intends to shift the focus from academic texts to the language of ideas used to navigate in disciplinary works, such as how the ideas and content can be expressed in various ways, and the linguistic resources learners of unprivileged backgrounds use to engage in the tasks. [ABSTRACT FROM AUTHOR]

This paper presents a study of emphasis in an attempt to determine its acoustic correlates in Central Mount Lebanon Lebanese Arabic (CMLL), an urban colloquial variety of Arabic spoken in Lebanon. Several analyses of emphasis exist in the literature, which largely focus on neighbouring vowels and either investigate, or make use of, a link between F2 lowering and emphasis. This study, based on acoustic data from 11 speakers, considers various possible consonantal as well as vocalic correlates. The vocalic analyses make use of statistical modelling in linear mixed effects, and generalised additive mixed modelling. This study finds significant F2 lowering in vowels preceding the emphatic voiceless sibilant /ṣ/. It argues that, in terms of the acoustics, emphasis is a vocalic rather than consonantal phenomenon, and that, in CMLL, the phonological feature 'emphasis' is consonantal, but is realized phonetically as a colouration of the adjacent vowels. [ABSTRACT FROM AUTHOR]

The poetry of Mexican author Jaime Sabines (1926–1999) is situated within a period of change in Latin America (Cuban revolution, dictatorships, and stylistic changes in literature) that linked the author to colloquialism. According to Cesare Del Mastro, this is a form marked by the use of plain and popular language rather than academicism. Using this style as a platform, two central topics presented in the collection of poems Diario Semanario y poemas en prosa (1961) are addressed in this paper: the everyday and the mortuary. Both are shown as symptomatic expressions of an affront to modernity, with signs of oppression, and the tense relationships and poetic responses are analyzed as a response to deep historical transformations. [ABSTRACT FROM AUTHOR]

The COVID pandemic has sparked fear among many people worldwide and has thus led to the emergence of a variety of conspiracy theories. Individuals believing in these theories come from various social and demographic backgrounds, some of them being mere skeptics, while others are more radical and extreme. The present paper investigates the use of language in a conspiratorial anti-COVID Facebook group with the aim of describing the linguistic features and strategies employed to share and spread conspiracy theories and to form a common identity. The results show that the group uses features typical of extremist groups, such the ideological in- and out-group presentation, the use of agentless passives to create fear, and intertextual references. Further strategies detected in the data are the misrepresentation of scientific knowledge, the use of colloquial language, humor, and irony, among others, which are used to create a bond among the groups' members and to set themselves apart from the government and the rest of the population, whom they perceive as enemies and oppressors. [ABSTRACT FROM AUTHOR]

In multi-lingual societies, where multiple languages are spoken in a small geographic vicinity, informal conversations often involve mix of languages. Existing speech technologies may be inefficient in extracting information from such conversations, where the speech data is rich in diversity with multiple languages and speakers. The DISPLACE (DIarization of SPeaker and LAnguage in Conversational Environments) challenge constitutes an open-call for evaluating and bench-marking the speaker and language diarization technologies on this challenging condition. To facilitate this challenge, a real-world dataset featuring multilingual, multi-speaker conversational far-field speech was recorded and distributed. The challenge entailed two tracks: Track-1 focused on speaker diarization (SD) in multilingual situations while, Track-2 addressed the language diarization (LD) in a multi-speaker scenario. Both the tracks were evaluated using the same underlying audio data. Furthermore, a baseline system was made available for both SD and LD task which mimicked the state-of-art in these tasks. The challenge garnered a total of 42 world-wide registrations and received a total of 19 combined submissions for Track-1 and Track-2. This paper describes the challenge, details of the datasets, tasks, and the baseline system. Additionally, the paper provides a concise overview of the submitted systems in both tracks, with an emphasis given to the top performing systems. The paper also presents insights and future perspectives for SD and LD tasks, focusing on the key challenges that the systems need to overcome before wide-spread commercial deployment on such conversations. • This paper presents an elaborated summary of the DISPLACE Challenge-2023. The challenge had two tracks: speaker diarization in multilingual settings (Track-1) and language diarization in multi-speaker scenarios (Track-2). • The DISPLACE dataset, comprising far-field multi-speaker multilingual conversational recordings, is released publicly as a part of this manuscript. The dataset also includes speaker labels, language labels, and transcriptions. • A detailed description of the DISPLACE corpus is presented, covering recording setup, annotation, and data analysis insights. • This paper details the baseline approaches and top three systems for both challenge tracks, highlighting error sources and suggesting future research avenues. [ABSTRACT FROM AUTHOR]

Research on the effects on early bilingualism on reading development is scarce. Here, we tackle this question by investigating whether the exposure to dual-language contexts (use of two languages in the same conversational situation) induces a boost in phonological and reading performance through more balanced spatial auditory attentional skills. Thirty Grade 1 early Spanish-Basque bilingual children predominantly exposed to dual-language contexts and 30 matched Spanish-Basque bilingual children predominantly exposed to single-language contexts were assessed on a dichotic listening task: syllables were simultaneously presented to the left and right ears and children were instructed to report the syllable they heard best. The dual-language contexts group demonstrated a reduced right ear advantage than the single-language contexts group, suggesting a more balanced report of syllables across ears in the former group. Moreover, more balanced reports were associated with more efficient reading skills. These results suggest that the frequent exposure to dual-language contexts possibly triggers the spatial rebalance of attentional resources for speech processing, impacting positively the development of reading skills. Future studies will determine whether some bilingual experience might act as a protective factor to reduce the prevalence and severity of future reading impairments such as dyslexia. [ABSTRACT FROM AUTHOR]

The advent of globalization and adaptation to multiple cultures has emanated a fusion of Hindi and English, casually known as Hinglish. The phenomenon of mixing multiple languages (such as Hindi and English) within a single utterance is often called code-mixing. Lately, code-mixed Hinglish has emerged as a dominant conversational language for Hindi-speaking citizens both online (on social media platforms) and offline. Although previous studies investigated such linguistic traits of Hinglish over the past few years, some pertinent questions still need to be answered: How did Hinglish evolve? And, what are the factors behind the evolution of Hinglish? Does the fusion of English impact all Hindi words similarly? To this end, we explore the empirical and statistical shreds of evidence behind the rise of Hinglish on social media such as Twitter. We show that adopting Hinglish depends on several socio-economic and demographic factors. We further formulate dynamic models to explore the socio-economic factors driving the growth of Hinglish, derive the future growth of Hinglish in the upcoming years, and estimate the propensity of users to change their linguistic preferences. Our study highlights that the Hinglish population has evolved steadily between 2014 and 2022, with an annualized growth rate of 1.2%, and the usage of Hinglish on Twitter has increased annually by 2%. Further, we find that the impact of Hinglish evolution is not uniform across different word groups and affects the contextual meaning of different words differently. Although our findings are specific to the Indian Hinglish community, our study can be generalized to understand the evolution and dynamics of other code-mixed languages, such as Spanish-English or Chinese-English. [ABSTRACT FROM AUTHOR]

Getting all students involved in a class can be a difficult task. Getting students excited about reading, writing, and learning can be an even more difficult task. Fortunately, using dialogues is an easy strategy that any teacher can implement in any grade in any subject area to get students up, moving, and actively engaged in the learning process. Dialogues are essentially two-person plays or skits that students act out in pairs. Each dialogue focuses on a specific topic and introduces key vocabulary terms and concepts in a way that students can understand, using day-to-day conversational language. In addition, all students love using dialogues because they appeal to various interests. The "science-y" students love the content, the lit/drama students love the format, and the class comics love the humor! AND they can be personalized and modified to meet the needs of specific students and educational goals and can be used with other strategies such as exit slips, labs, content review, and more. In short, dialogues are an easy way to create an inclusive classroom where every student is engaged and successful. [ABSTRACT FROM AUTHOR]

In the rapidly evolving domain of conversational agents, the integration of Large Language Models (LLMs) into Chatbot Development Platforms (CDPs) is a significant innovation. This study compares the efficacy of employing generic and fine-tuned GPT-3.5-turbo models for designing dialog flows, focusing on the intent and entity recognition crucial for dynamic conversational interactions. Two distinct approaches are introduced: a generic GPT-based system (G-GPT) leveraging the pre-trained model with complex prompts for intent and entity detection, and a fine-tuned GPT-based system (FT-GPT) employing customized models for enhanced specificity and efficiency. The evaluation encompassed the systems' ability to accurately classify intents and recognize named entities, contrasting their adaptability, operational efficiency, and customization capabilities. The results revealed that, while the G-GPT system offers ease of deployment and versatility across various contexts, the FT-GPT system demonstrates superior precision, efficiency, and customization, although it requires initial training and dataset preparation. This research highlights the versatility of LLMs in enriching conversational features for talking assistants, from social robots to interactive chatbots. By tailoring these advanced models, the fluidity and responsiveness of conversational agents can be enhanced, making them more adaptable and effective in a variety of settings, from customer service to interactive learning environments. [ABSTRACT FROM AUTHOR]

The study of biological materials and bio‐inspired materials science is well established; however, surprisingly little knowledge is systematically translated to engineering solutions. To accelerate discovery and guide insights, an open‐source autoregressive transformer large language model (LLM), BioinspiredLLM, is reported. The model is finetuned with a corpus of over a thousand peer‐reviewed articles in the field of structural biological and bio‐inspired materials and can be prompted to recall information, assist with research tasks, and function as an engine for creativity. The model has proven that it is able to accurately recall information about biological materials and is further strengthened with enhanced reasoning ability, as well as with Retrieval‐Augmented Generation (RAG) to incorporate new data during generation that can also help to traceback sources, update the knowledge base, and connect knowledge domains. BioinspiredLLM also has shown to develop sound hypotheses regarding biological materials design and remarkably so for materials that have never been explicitly studied before. Lastly, the model shows impressive promise in collaborating with other generative artificial intelligence models in a workflow that can reshape the traditional materials design process. This collaborative generative artificial intelligence method can stimulate and enhance bio‐inspired materials design workflows. Biological materials are at a critical intersection of multiple scientific fields and models like BioinspiredLLM help to connect knowledge domains. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to explore the role of the linguistic style used in the brand-posted social media content on consumer engagement in the Fintech domain. Design/methodology/approach: A total of 3,286 tweets (registering nearly 1.35 million impressions) published by 10 leading Fintech unicorns in India were extracted using the Twitter API. The Linguistic Inquiry and Word Count (LIWC) dictionary was used to analyse the linguistic characteristics of the shared tweets. Negative Binomial Regression (NBR) was used for testing the hypotheses. Findings: This study finds that using drive words and cognitive language increases consumer engagement with Fintech messages via the central route of information processing. Further, affective words and conversational language drive consumer engagement through the peripheral route of information processing. Research limitations/implications: The study extends the literature on brand engagement by unveiling the effect of linguistic features used to design social media messages. Practical implications: The study provides guidance to social media marketers of Fintech brands regarding what content strategies best enhance consumer engagement. The linguistic style to improve online consumer engagement (OCE) is detailed. Originality/value: The study's findings contribute to the growing stream of Fintech literature by exploring the role of linguistic style on consumer engagement in social media communication. The study's findings indicate the relevance of the dual processing mechanism of elaboration likelihood model (ELM) as an explanatory theory for evaluating consumer engagement with messages posted by Fintech brands. [ABSTRACT FROM AUTHOR]

With this article I explore the relationship between core academic language skills (CALS) - a construct representing a set of high-utility language skills needed to manage the linguistic features prevalent in academic texts across content areas - and schooling outcomes. There has long been an understanding that there is a distinction between academic language and colloquial language, originally described by Jim Cummins (1976) as cognitive academic language proficiency (CALP) and basic interpersonal communication skills (BICS). The construct has only recently been operationalised (Uccelli, Phillips Galloway, Barr, Meneses & Dobbs, 2015) as the individual skills and competencies that underlie CALP. The CALS construct describes an empirically testable set of competencies that address this need. This has been used in the development of an assessment instrument aimed at South African learners - the CALS-I-ZA (MacFarlane, Barr & Uccelli, 2022) - and I investigate whether a measurable link exists between this assessment and schooling outcomes for a sample of Grade 6 learners in 2 public schools in the Gauteng province of South Africa. Schooling outcomes have been measured using the Gauteng Provincial Common Assessments - a provincial examination intended to measure schooling outcomes on a comparable assessment instrument. The study reported on here shows a moderately strong correlation between the CALS-I-ZA and the provincial common examinations (r = 0.64 and r = 0.65). This predictive relationship between CALS and schooling outcomes leads to an argument for direct instruction in CALS as an embedded feature of pedagogy in South Africa. [ABSTRACT FROM AUTHOR]

Capture land is a Jamaican colloquialism for land that is (or is presumed to be) occupied without the authorization of the landowner. Working through two quotes, "just to make life for a time," and, "we've always been here," this article examines how the practice of capture indexes abolitionist spatial practices of varied temporalities: rooted in belonging to place, but also alighting provisionally. Through ethnographic research, this article reveals the use of capture for both placemaking and flight, parallel but in tension with the fixity and fluidity inhered in the liberal property regime. Compared with earlier ethnographies of land tenure in Jamaica, this points to an epistemology that is less preoccupied by ownership. What takes primacy is the care for generations of plants, children, rebels, and futures unfolding through "the plot" and its multiple agendas—sustenance (plot of land), refusal (plot to rebel) and the plotting of alternate futures. While here emerges through the spatiotemporalities of life in the interstices of racial capitalism, it also challenges us to reimagine how to think land after colonialism and beyond property. From here, we might imagine simultaneity of apparently discrepant temporalities of Black freedom dreams and speculate on the spacetimes of "freedom is a place." [ABSTRACT FROM AUTHOR]

This study was carried out with the main objective of investigating the impact of the English language lexeme on modern colloquial Sinhala and exploring the specific mechanisms of neologism and borrowing. Among the major findings, the impact of English on colloquial Sinhala has been identified in the processes of the acronym, coinage, alphabetic abbreviation, clipping, blending, generification, and borrowing as the dominant processes that come under the umbrella term of neologism. As borrowing mechanisms, loan words, loan shifts and loan blends followed by other subsequent mechanisms like loan translations, loan renditions and loan creations have been established with colloquial Sinhala. Participatory observation, chunk recordings and secondary data sources are methodological instruments of this research. The Englishinfluenced lexical items in modern colloquial Sinhala have been critically analysed and categorized logically with the final concluding remark that the English language's influence on modern colloquial Sinhala is inordinate with special reference to lexical categories. [ABSTRACT FROM AUTHOR]

This research tried to investigate a specific case of pronominal enclitic (eš/aš) in colloquial Persian based on functional approach. Two novels including a total of 56,500 words and the data on the internet were used as corpus. A corpus of 110 sentences with the desired structure were extracted. The investigation revealed that the reference of pronominal enclitic is the common and assumed knowledge of interlocutors which might be represented as a proposition in the discourse or activated by an element in the language. The pronominal enclitic is structurally a presupposition trigger and adjuncts (rastaš, haqiqataš, vaqeiataš) are discourse markers. The reference enclitic holds old information and is a specified definite. The predicate following enclitic holds new information and is a specified indefinite. This predicate contains focal information. The third fason singular enclitic appears both as an adjunct and as a clause at the beginning of the sentence and can be omitted. Their semantic and pragmatic function is truth assertion. [ABSTRACT FROM AUTHOR]

Copyright of Analisi: Quaderns de Comunicacio i Cultura is the property of Universitat Autonoma de Barcelona and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

For proper assessment, diagnosis, and treatment of language disorders in Mandarin children, it is important to have measures that closely track the course of normal development. The current study uses a large collection of spontaneous conversational language samples to track the developmental course of five language measures: mean length of utterance (MLU), the MLU of the five longest utterances (MLU5), vocD, number of repetitions, and number of retracings. We used cross-validation-based linear regression to estimate the relationship between age, gender and each of the five variables derived from the conversational language samples in 101 typically developing Mandarin-speaking children aged 3 to 7. Each of the five measures showed significant age-related effects during the period from age 3 to age 7. As norm-referenced measures of language development in children speaking Mandarin, these developmental data could inform clinical therapists regarding assessments and interventions for children with language disorder or impairment. [ABSTRACT FROM AUTHOR]

The document is a review of a book titled "Usage-based and Typological Approaches to Linguistic Units." The book examines whether linguistic categories like subject, clause, or predicate are universal across languages or specific to each language. It presents research articles that use authentic corpus data and a usage-based approach to explore the universality of linguistic units and the concept of "participant orientation" in different languages. The review discusses the perspectives and findings presented in the book, including studies on English, Finnish, Japanese, and Indonesian conversations. The reviewer suggests future research should include more languages, consider diachronic analysis, and explore the role of prosody in linguistic units. Overall, the book challenges traditional views on the universality of linguistic categories and provides valuable insights into the study of linguistic units. It is recommended for those interested in gaining a deeper understanding of these topics. [Extracted from the article]

This essay delves into the multifaceted nature of the Tagalog word ano, examining its usage in various contexts and its impact on communication within Tagalog-speaking communities. Three cases of ano are discussed in the essay. The first case pertains to ano accompanied by ostension; the second, pertains to ano accompanied by context clues; and the third, pertains to ano without any ostension or context. In Tagalog communities, ano, despite the lack of ostension or context, is sometimes still used in everyday conversations, presenting the problem of how meaning or understanding of the word ano is still possible. The third case of ano will be further divided into two: IT-ano (ibang tao-ano) and HIT-ano (hindi ibang tao-ano) providing the difference between strangers and acquaintances. By relying on Ludwig Wittgenstein's idea of what it means to understand through language-game, this essay discusses how a vague word such as ano gives rise to meaning and understanding. This approach to clarify meaning will relate to the levels of social interaction found in Virgilio Enriquez's idea of pakikipagkapwa. The analysis will proceed by referring to hypothetical scenarios involving the use of the term ano situated in the everyday conversational use of language by Filipinos. The mode of analysis takes the perspective of the author, who is a native speaker of Tagalog. [ABSTRACT FROM AUTHOR]