Language Text Research Articles

Automatically Generated Texts for Fauna and Flora from Structured Data Based on a Controlled Vocabulary

Information systems like Xper3 and Fulgoromorpha Lists On the Web (FLOW) play a crucial role in managing and using biological data. These platforms store extensive collections of normalized data and structured taxonomic descriptions. By using controlled terminologies, they standardize the vocabulary, significantly enhancing the processes of identification, description, and comparison of various taxa. The massive assemblages of data hosted in these repositories could be reused to generate texts in natural languages automatically. The most immediate goal is to produce, from these information systems, more accessible and user-friendly displays in the form of taxon summary pages. This automated production of textual outputs is a great addition that can be continuously updated as the databases evolve. Can structured data be reused to provide better species pages and to ensure updating if the data evolves? Will AI assist in this process, or will specific computing be needed? To address these questions, multiple possible approaches have been identified. The most basic level of this process involves converting a single line from a taxon-by-character matrix into text that resembles natural language, similar to the descriptions found in botanical or zoological guides. The primary goal is to move beyond the rigid format of matrix lines or lists of characteristics (such as characters and states) of a species, and instead generate a coherent, easy-to-read paragraph intended for human eyes. To achieve this objective, a first solution is given by Descrxp, a tool currently developed in conjunction with Xper. The user specifies the desired key outline for the output, and Descrxp fills this canvas using the database contents. While this approach is highly reliable and adaptable to various contexts, its drawback is being labor-intensive, and requiring significant human input and oversight (Fig. 1). The second solution uses AI. Data on well-known taxa, and on more obscure taxa have been tested with ChatGPT 3.5 or 4.0. ChatGPT succeeds well at generating natural language descriptions for well-known taxa (Fig. 2), providing text that appears accurate and coherent due to the extensive information available online for these groups. However, when it comes to more obscure taxa, less represented on the internet, the results can be more inconsistent and unpredictable (Fig. 3). This variability can be explained by the quality of the data the model has been trained on, which is significantly richer for well-known species. Consequently, for lesser-known groups, the generated descriptions may lack the precision and reliability seen with more familiar taxa. However, if more careful attention is given to the output of chatGPT (Fig. 4), it is possible to notice that the AI filled part of the description with inaccurate characters. Due to its inability to recognize specific scientific terminology, and to process hierarchical information, chatGPT ends up producing an over-generalized and redundant description. To go further, it would be possible to use a comparison between taxa to produce a text that highlights the remarkable features of a taxon. Moreover, it could be interesting to try and synthesize several lines of the initial matrix to extract generalized descriptions of groups of taxa, such as the description of a genus based on its included species. Using data from FLOW, Fishbase, and Xper3, the French ACDC (Counterfactual Learning for Controlled Data-to-text) project is creating a variety of text outputs. These range from "fill-in-the-blank" canvas to fully AI-generated content, utilizing data matrices derived from these pilot databases.

Membership categorisation, sociological description and role prompt engineering with ChatGPT

Large Language Models (LLMs) and generative Artificial Intelligence (A.I.) have become the latest disruptive digital technologies to breach the dividing lines between scientific endeavour and public consciousness. LLMs such as ChatGPT are platformed through commercial providers such as OpenAI, which provide a conduit through which interaction is realised, via a series of exchanges in the form of written natural language text called ‘prompt engineering’. In this paper, we use Membership Categorisation Analysis to interrogate a collection of prompt engineering examples gathered from the endogenous ranking of prompting guides hosted on emerging generative AI community and practitioner-relevant social media. We show how both formal and vernacular ideas surrounding ‘natural’ sociological concepts are mobilised in order to configure LLMs for useful generative output. In addition, we identify some of the interactional limitations and affordances of using role prompt engineering for generating interactional stances with generative AI chatbots and (potentially) other formats. We conclude by reflecting the consequences of these everyday social-technical routines and the rise of ‘ethno-programming’ for generative AI that is realised through natural language and everyday sociological competencies.

Use of Language By generative AI Tools in Mathematical Problem Solving: The Case of ChatGPT

Texts generated by artificial intelligence agents have been suggested as tools supporting students’ learning. The present research analyses the language of texts generated by ChatGPT when solving mathematical problems related to the quadratic equation. We use the functional grammar theoretical framework that includes three meta-functions: the ideational meta-function, the interpersonal meta-function and the textual meta-function. The results indicated that in at least one of six problem-solving tasks ChatGPT provided a mathematically incorrect answer. The processes appearing in ChatGPT texts, aiming at developing students’ understanding of mathematical concepts, included verbal, mental, existential, relational and behavioural processes but no material processes. Specifically, ChatGPT performed a mathematically incorrect existential process. ChatGPT generally used the first plural pronoun ‘we’ when describing the processes of solving mathematical problems, while it generally used the first-person singular pronoun when taking responsibility for a specific mistake or when expressing happiness for the actions of the user. Moreover, generally the text of the solution did not include direct imperatives but used ‘let us do’. The advancement of the ChatGPT textual solution was made usually through steps like ‘first’, ‘second’, etc. The research results indicated that the way ChatGPT responded to the mathematical problems would be useful in supporting learners’ understanding of ways to solve quadratic equations, but only if the teacher critically accompanies the student in the problem-solving process. Self-study with ChatGPT could lead to or confirm students’ mathematical misconceptions.

IT Applications for the Tuvan Language: Regional Practices and Future Plans

In the era of digitalization and globalization, the likelihood of the disappearance of small languages, including the Tuvan language is very high. One of the reliable and promising solutions for preserving a language is its inclusion in the digital space as a language of accumulation, processing and transmission of information, a language of communication on the Internet. Currently, a number of tasks aimed at preserving and popularizing the Tuvan language are being car­ried out within the framework of the state program to support the development of the Tuvan language in the Republic of Tuva. Various government agencies of the Republic of Tuva and interested IT specialists are working on the digitalization of the Tuvan language. Tuvan State University (TuvSU) also makes a significant contribution to the development and popularization of the Tuvan language. Teachers of the Department of Informatics of TuvSU together with scientists from the Faculty of Philology and the "Turkology" scientific and educational centerwork in the following areas: development of software products for teaching the Tuvan language, creation of software tools for text processing in the Tuvan language, creation of linguistic information resources. Along with TuvSU teachers, students from the faculties of physics, mathematics and philology also take part in this research work. The topics of students' coursework and final qualification papers are related to IT developments. This article presents the experience of research and development of teachers of the Department of Computer Science in the field of preservation, study and development of the Tuvan language using information and communication technologies in these three areas. In the first direction, electronic textbooks, interactive simulators, and mobile applications for teaching the Tuvan language are being developed. The second direction is focused on the creation of software products for processing texts in the Tuvan language — a morphological analyzer of the Tuvan language, program for checking Tuvan texts for plagiarism. The last direction is related to the creation of electronic dictionaries and translators for the Tuvan language, which can be used in all spheres of life.

Targeted Training Data Extraction—Neighborhood Comparison-Based Membership Inference Attacks in Large Language Models

A large language model refers to a deep learning model characterized by extensive parameters and pretraining on a large-scale corpus, utilized for processing natural language text and generating high-quality text output. The increasing deployment of large language models has brought significant attention to their associated privacy and security issues. Recent experiments have demonstrated that training data can be extracted from these models due to their memory effect. Initially, research on large language model training data extraction focused primarily on non-targeted methods. However, following the introduction of targeted training data extraction by Carlini et al., prefix-based extraction methods to generate suffixes have garnered considerable interest, although current extraction precision remains low. This paper focuses on the targeted extraction of training data, employing various methods to enhance the precision and speed of the extraction process. Building on the work of Yu et al., we conduct a comprehensive analysis of the impact of different suffix generation methods on the precision of suffix generation. Additionally, we examine the quality and diversity of text generated by various suffix generation strategies. The study also applies membership inference attacks based on neighborhood comparison to the extraction of training data in large language models, conducting thorough evaluations and comparisons. The effectiveness of membership inference attacks in extracting training data from large language models is assessed, and the performance of different membership inference attacks is compared. Hyperparameter tuning is performed on multiple parameters to enhance the extraction of training data. Experimental results indicate that the proposed method significantly improves extraction precision compared to previous approaches.

Sign language literary translations: linguistic, dramatic and technological aspects

The research reported here considers how norms of Libras (Brazilian Sign Language) literature (Sutton-Spence, 2021) operate with deaf translation norms (Stone, 2005) and literary translation norms (Toury, 1995) to create some potential norms of deaf literary translation. Presenting some considerations upon the concept of the “visual” in Libras literature and translation, we draw upon translations in three different genres of literature, made by three deaf translators, to understand better the key multimodal relationships between signed literature, written texts, images and video editing in literary translation. Analyzed in terms of the linguistic, dramatic and technological elements (Bartolomei; Perreira, 2021), we show how the translations create acceptable target language texts by focusing on the visual esthetics of the texts.

Natural Language Processing Pretraining Language Model for Computer Intelligent Recognition Technology

Computer intelligent recognition technology refers to the use of computer vision, Natural Language Processing (NLP), machine learning and other technologies to enable computers to recognize, analyze, understand and answer human language and behavior. The common applications of computer intelligent recognition technology include image recognition, NLP, face recognition, target tracking, and other fields. NLP is a field of computer science, which involves the interaction between computers and natural languages. NLP technology can be used to process, analyze and generate natural language data, such as text, voice and image. Common NLP technology applications include language translation, emotion analysis, text classification, speech recognition and question answering system. Language model is a machine learning model, which uses a large number of text data for training to learn language patterns and relationships in text data. Although the language model has made great progress in the past few years, it still faces some challenges, including: poor semantic understanding, confusion in multilingual processing, slow language processing and other shortcomings. Therefore, in order to optimize these shortcomings, this article would study the pre-training language model based on NLP technology, which aimed at using NLP technology to optimize and improve the performance of the language model, thus optimizing the computer intelligent recognition technology. The model had a higher language understanding ability and more accurate prediction ability. In addition, the model could learn language rules and structures by using a large number of corpus, so as to better understand natural language. Through experiments, it could be known that the data size and total computing time of the traditional Generative Pretrained Transformer-2 (GPT-2) language model were 10 GB and 97 hours respectively. The data size and total computing time of BERT (Bidirectional Encoder Representations from Transformer) were 12 GB and 86 hours respectively. The data size and total computing time of the pre-training language model based on NLP were 18 GB and 71 hours respectively. Obviously, the pre-training language model based on NLP had a larger data size and shorter computing time. The experimental results showed that the NLP technology could better optimize the language model and effectively improve its various capabilities. This article opened up a new development direction for computer intelligent recognition technology and provided excellent technical support for the development of language models.

Tinjauan Kedwibahasaan Buku Cerita Anak ‘The Foos’ dari Pandangan Metode Penerjemahan

Bilingual children's story books in Indonesian and English are now being published in Indonesia, providing them as reading options for mothers, caregivers, storytellers, and the children themselves. This study aims to analyze, identify, describe, and classify the forms of translation methods found in a bilingual children's storybook entitled 'The Foos' by Arleen Amidjaja, as well as to analyze the translation methods applied to determine the success in achieving equivalence of meaning in the target language (TL). This study uses a qualitative descriptive method and a documentation method. Employing text analysis, careful reading, and note taking techniques for research constitutes the stages of data collection. Text analysis uses Newmark's translation method theory with data sources from English texts as the source language text (SL) and Indonesian as the TL from the storybook 'The Foos.' The results of the analysis are presented in the form of a comparison of the two languages ​​, along with the results of identifying SL and TL text translation techniques. The interpretation of the identification results is then presented in the form of a descriptive presentation by considering the translation techniques of the identification results. The results of this study indicate that 12 translation method data are oriented to the SL, and eight are oriented to the TL. The translation techniques that are often used are semantic and communicative translation techniques. There are no faithful, adaptation, or idiomatic translation techniques because the children's storybook 'The Foos' does not have proverbial sentences or other foreign language terms. The SL uses simple sentence structures and vocabulary. The translation method that is oriented to the SL with semantic and communicative translation techniques is effective for translating children's storybooks, ensuring that readers understand the content of the stories.

The Idea of a “Flawed Treaty” and its Influence on Students’ Perceptions of the New Zealand Wars

AbstractThis article reports on students’ accounts of te Tiriti o Waitangi as a flawed agreement that was a cause of subsequent conflict. We examine how the notion of a “flawed treaty” is developed in history and educational texts. We argue that when the cause of conflict is attributed to the failures of those engaged in the historical production of the treaty documents, Māori perspectives on the sacred significance of the treaty are obscured, as is the contemporary relevance of the agreement. As much as possible, throughout this article, the capitalised phrase te Tiriti o Waitangi (or te Tiriti) is used to refer especially to the te reo Māori text and the Treaty of Waitangi (or the Treaty) to the English language text of the treaty. ‘The treaty’ (in lower case) refers to an agreement to an ongoing relationship, and where a specific text is not defined.

HU U?: The Impact of Texting Language on the Academic Writing Proficiency of Second Language Learners

This study examined the common language used for texting between first- and second-year English language studies majors at a state university in the northern Philippines, as well as their writing skills. The methodology of total enumeration sampling was utilized to ascertain the study participants. The findings demonstrated that "Shortenings and Abbreviations" is the most often used texting language among the seven commonly used texting languages among the respondents. The results also revealed that the respondents' written performance had a relatively low level of writing competence, "Developing," which can be linked to their usage of acronyms and shorthands. These imply that the idea that textese, or texting language, is destroying language and influencing the language performance of the students has been proven and confirmed.

Easy-read and large language models: on the ethical dimensions of LLM-based text simplification

The production of easy-read and plain language is a challenging task, requiring well-educated experts to write context-dependent simplifications of texts. Therefore, the domain of easy-read and plain language is currently restricted to the bare minimum of necessary information. Thus, even though there is a tendency to broaden the domain of easy-read and plain language, the inaccessibility of a significant amount of textual information excludes the target audience from partaking or entertainment and restricts their ability to live life autonomously. Large language models can solve a vast variety of natural language tasks, including the simplification of standard language texts to easy-read or plain language. Moreover, with the rise of generative models like GPT, easy-read and plain language may be applicable to all kinds of natural language texts, making formerly inaccessible information accessible to marginalized groups like, a.o., non-native speakers, and people with mental disabilities. In this paper, we argue for the feasibility of text simplification and generation in that context, outline the ethical dimensions, and discuss the implications for researchers in the field of ethics and computer science.

A universal knowledge model and cognitive architectures for prototyping AGI

The article identified 56 cognitive architectures for creating general artificial intelligence (AGI) and proposed a set of interrelated functional blocks that an agent approaching AGI in its capabilities should possess. Since the required set of blocks is not found in any of the existing architectures, the article proposes a reference cognitive architecture for intelligent systems approaching AGI in their capabilities. As one of the key solutions within the framework of the architecture, a universal method of knowledge representation is proposed, which allows combining various non-formalized, partially and fully formalized methods of knowledge representation in a single knowledge base, such as texts in natural languages, images, audio and video recordings, graphs, algorithms, databases, neural networks, knowledge graphs, ontologies, frames, essence-property-relation models, production systems, predicate calculus models, conceptual models, and others. To combine and structure various fragments of knowledge, archigraph model are used, constructed as a development of annotated metagraphs. As other components, the reference cognitive architecture being developed includes following modules: machine consciousness, machine subconsciousness, interaction with the external environment, a goal management, an emotional control, social interaction, reflection, ethics, worldview, learning, monitoring, statement problems, solving problems, self-organization and meta learning. Based on the composition of the proposed reference architecture modules, existing cognitive architectures containing the following modules were analyzed: machine consciousness, machine subconsciousness, reflection, worldview.

ZeroNLG: Aligning and Autoencoding Domains for Zero-Shot Multimodal and Multilingual Natural Language Generation.

Natural Language Generation (NLG) accepts input data in the form of images, videos, or text and generates corresponding natural language text as output. Existing NLG methods mainly adopt a supervised approach and rely heavily on coupled data-to-text pairs. However, for many targeted scenarios and for non-English languages, sufficient quantities of labeled data are often not available. As a result, it is necessary to collect and label data-text pairs for training, which is both costly and time-consuming. To relax the dependency on labeled data of downstream tasks, we propose an intuitive and effective zero-shot learning framework, ZeroNLG, which can deal with multiple NLG tasks, including image-to-text (image captioning), video-to-text (video captioning), and text-to-text (neural machine translation), across English, Chinese, German, and French within a unified framework. ZeroNLG does not require any labeled downstream pairs for training. During training, ZeroNLG (i) projects different domains (across modalities and languages) to corresponding coordinates in a shared common latent space; (ii) bridges different domains by aligning their corresponding coordinates in this space; and (iii) builds an unsupervised multilingual auto-encoder to learn to generate text by reconstructing the input text given its coordinate in shared latent space. Consequently, during inference, based on the data-to-text pipeline, ZeroNLG can generate target sentences across different languages given the coordinate of input data in the common space. Within this unified framework, given visual (imaging or video) data as input, ZeroNLG can perform zero-shot visual captioning; given textual sentences as input, ZeroNLG can perform zero-shot machine translation. We present the results of extensive experiments on twelve NLG tasks, showing that, without using any labeled downstream pairs for training, ZeroNLG generates high-quality and "believable" outputs and significantly outperforms existing zero-shot methods.

Pāli as the Language of Buddhist Scripture

This research aims to analytically examine the Pāli language as the language of the Buddhist sacred texts. This research is carried out by library method with content analysis as its primary approach. The data are gathered from various literary sources such as journals, books and relevant articles related to the research topic. As result, the chronology of the Pāli as the sacred language of Buddhist text can be understood. In early period, the term Pāli did not indicate as a language, but it refered to the texts or Buddhist scripture. Pāli is defined as the texts or scripture (pāṭhe iti pi pāḷi) which means “that which protects.” It is called Pāli because it protects the teaching (pāḷīti atthaṃ pāletīti pāḷi). Therefore, the term Pāli is considered synonymous with terms Buddhavacana, Tipiṭaka, Tanti, and Pariyatti. The language had been used by the Buddha when delivering the teachings was identified as Māgadhī. Pāli literature such as commentaries and chronicles hightlight that Pāli as Māgadhī, Māgadhānirutti or Māgadhikabhāsā which was the language spoken by the Buddha during his lifetime. In conclusion, at the beginning Pāli means the scripture, but later it came to be recognized as the language of the scripture.

The Difficulties of Students’ Understanding on Figurative Language and Feasible Suggestions

Figurative language of creativity is common in our daily life. It is due to the ubiquity of it in written and spoken texts that a second language learner is likely to inevitably encounter all kinds of figurative expressions. Learners of English who come from different cultural background are likely to have great difficulties in understanding figurative language in texts. Based on American former president Trump’s speech, this essay identifies the figurative language which could pose difficulties for a learner of English in the text, especially focused on learners from China and explains why learners may experience these difficulties. The essay suggests that using the context, literal meaning and corpus lines may be helpful for learners to aid comprehension and overcome metaphor difficulties.

Translating Japanese Wordplay into English and Filipino A Linguistic Analysis of Joke Translation in Spy x Family Subtitles

Japanese animated shows called “anime” have seen increased popularity among Philippine audiences in the 2000s and beyond partly because of television broadcasts and online streaming platforms, but one possible hindrance to the understanding of the media is the language gap, since Japanese and the Philippine national lingua franca Filipino, along with the international lingua franca English, are unrelated languages. To bridge this gap, a number of channels and sites provide translations of the shows into th e languages spoken by Filipinos, and one of the forms that they take are subtitles: texts in the target language placed near the bottom of the screen, providing viewers with line-by-line translations of the dialogue. Through the analysis of three examples of Japanese linguistic humor and their corresponding translations into Filipino and English, this paper aimed to understand strategies translators use in presenting wordplay to unrelated languages. Results reveal that a word – by-word translation is not always followed, and translators put the target languages’ culture, structure, lexicon and patterns into consideration. Examples include the use of the figurative-literal dichotomy of the meanings of words in both Filipino and English when the original Japa nese material utilizes the exactly similar pronunciation between two words with entirely different meanings. Keywords : sociolinguistics, multilingualism, globalization, translation

Privacy Preservation of Large Language Models in the Metaverse Era: Research Frontiers, Categorical Comparisons, and Future Directions

ABSTRACTLarge language models (LLMs), with their billions to trillions of parameters, excel in natural language processing, machine translation, dialog systems, and text summarization. These capabilities are increasingly pivotal in the metaverse, where they can enhance virtual interactions and environments. However, their extensive use, particularly in the metaverse's immersive platforms, raises significant privacy concerns. This paper analyzes existing privacy issues in LLMs, vital for both traditional and metaverse applications, and examines protection techniques across the entire life cycle of these models, from training to user deployment. We delve into cryptography, embedding layer encoding, differential privacy and its variants, and adversarial networks, highlighting their relevance in the metaverse context. Specifically, we explore technologies like homomorphic encryption and secure multiparty computation, which are essential for metaverse security. Our discussion on Gaussian differential privacy, Renyi differential privacy, Edgeworth accounting, and the generation of adversarial samples and loss functions emphasizes their importance in the metaverse's dynamic and interactive environments. Lastly, the paper discusses the current research status and future challenges in the security of LLMs within and beyond the metaverse, emphasizing urgent problems and potential areas for exploration.

Of Shadowcats and Pipeweed: Fantastic Animal and Plant Names in a Cross-linguistic Comparison

The paper focuses on the names of fictitious species of animals and plants and on their position in the creation of fictional worlds. The examined sample of occasionalisms, or nonce words, is drawn from popular and influential writings, which include J. R. R. Tolkien’s works, Terry Pratchett’s texts, J. K. Rowling’s Harry Potter series, Suzanne Collins’s The Hunger Games series and George R. R. Martin’s The Song of Ice and Fire series. The position of the sign in the word-formation system of both the original language of the work and of selected translations is analyzed – mainly English, German, and Czech versions are taken into account. A cross-linguistic comparison leads to the conclusion that the choices made by the writer or the translator in naming fictional species correspond, on the one hand, to the linguistic habits of the language in question, while, on the other hand, providing room for linguistic experimentation and influence from the language of the original text. The paper also shows how the chosen texts, through the usage of fictitious names of species, attempt to meet the readers’ expectations for speculative literature. The presence of boundary crossing, hybridization, and linguistic creativity makes foregrounded world-building possible. However, accentuated connections between the words and between the species of the fictional and the actual world also show that an unambiguous, simpler world is presented this way. Word-formation is presented by the study as a way to link the actual world and the simplified speculative world, thus highlighting its importance.

Applying the Markov chain in natural language processing and three-pool model

This paper delves into the application of Markov chains in Natural Language Processing (NLP), and the Markov Chain Monte Carlo (MCMC) methodology relevant to the three-pool model. The former outlines the basic principles of Markov chains, highlighting their utility in predicting word sequences in language modelling and text generation, despite certain limitations. Also, the former describes mathematical frameworks like n-gram models that enhance prediction accuracy by considering multiple preceding words. It acknowledges challenges in NLP such as oversimplification and emotional depth, as well as computational issues in higher-order models. It concludes by discussing the integration of Markov chains with other models to mitigate these limitations, and their enduring relevance in computational linguistics. The later investigates the MCMC methodology, a seminal development in the field of statistical inference, which is especially useful when analysing complicated systems when traditional statistical techniques are inadequate. Moreover, this later explores the fundamental concepts of MCMC, clarifies how it is inherently related to Markov chains, presents the three-pool model that is commonly applied to models of physical, chemical, or ecological systems, and discusses how MCMC can be used to analyse these models.

App review driven collaborative bug finding

Software development teams generally welcome any effort to expose bugs in their code base. In this work, we build on the hypothesis that mobile apps from the same category (e.g., two web browser apps) may be affected by similar bugs in their evolution process. It is therefore possible to transfer the experience of one historical app to quickly find bugs in its new counterparts. This has been referred to as collaborative bug finding in the literature. Our novelty is that we guide the bug finding process by considering that existing bugs have been hinted within app reviews. Concretely, we design the BugRMSys approach to recommend bug reports for a target app by matching historical bug reports from apps in the same category with user app reviews of the target app. We experimentally show that this approach enables us to quickly expose and report dozens of bugs for targeted apps such as Brave (web browser app). BugRMSys ’s implementation relies on DistilBERT to produce natural language text embeddings. Our pipeline considers similarities between bug reports and app reviews to identify relevant bugs. We then focus on the app review as well as potential reproduction steps in the historical bug report (from a same-category app) to reproduce the bugs. Overall, after applying BugRMSys to six popular apps, we were able to identify, reproduce and report 20 new bugs: among these, 9 reports have been already triaged, 6 were confirmed, and 4 have been fixed by official development teams.