Hasil untuk "Language and Literature"

Su Lin Blodgett, Solon Barocas, Hal Daum'e et al.

We survey 146 papers analyzing “bias” in NLP systems, finding that their motivations are often vague, inconsistent, and lacking in normative reasoning, despite the fact that analyzing “bias” is an inherently normative process. We further find that these papers’ proposed quantitative techniques for measuring or mitigating “bias” are poorly matched to their motivations and do not engage with the relevant literature outside of NLP. Based on these findings, we describe the beginnings of a path forward by proposing three recommendations that should guide work analyzing “bias” in NLP systems. These recommendations rest on a greater recognition of the relationships between language and social hierarchies, encouraging researchers and practitioners to articulate their conceptualizations of “bias”---i.e., what kinds of system behaviors are harmful, in what ways, to whom, and why, as well as the normative reasoning underlying these statements—and to center work around the lived experiences of members of communities affected by NLP systems, while interrogating and reimagining the power relations between technologists and such communities.

Moin Nadeem, Anna Bethke, Siva Reddy

A stereotype is an over-generalized belief about a particular group of people, e.g., Asians are good at math or African Americans are athletic. Such beliefs (biases) are known to hurt target groups. Since pretrained language models are trained on large real-world data, they are known to capture stereotypical biases. It is important to quantify to what extent these biases are present in them. Although this is a rapidly growing area of research, existing literature lacks in two important aspects: 1) they mainly evaluate bias of pretrained language models on a small set of artificial sentences, even though these models are trained on natural data 2) current evaluations focus on measuring bias without considering the language modeling ability of a model, which could lead to misleading trust on a model even if it is a poor language model. We address both these problems. We present StereoSet, a large-scale natural English dataset to measure stereotypical biases in four domains: gender, profession, race, and religion. We contrast both stereotypical bias and language modeling ability of popular models like BERT, GPT-2, RoBERTa, and XLnet. We show that these models exhibit strong stereotypical biases. Our data and code are available at https://stereoset.mit.edu.

V. Tshitoyan, John Dagdelen, Leigh Weston et al.

M. Tremmel, U. Gerdtham, P. Nilsson et al.

Background: The rising prevalence of obesity represents an important public health issue. An assessment of its costs may be useful in providing recommendations for policy and decision makers. This systematic review aimed to assess the economic burden of obesity and to identify, measure and describe the different obesity-related diseases included in the selected studies. Methods: A systematic literature search of studies in the English language was carried out in Medline (PubMed) and Web of Science databases to select cost-of-illness studies calculating the cost of obesity in a study population aged ≥18 years with obesity, as defined by a body mass index of ≥30 kg/m², for the whole selected country. The time frame for the analysis was January 2011 to September 2016. Results: The included twenty three studies reported a substantial economic burden of obesity in both developed and developing countries. There was considerable heterogeneity in methodological approaches, target populations, study time frames, and perspectives. This prevents an informative comparison between most of the studies. Specifically, there was great variety in the included obesity-related diseases and complications among the studies. Conclusions: There is an urgent need for public health measures to prevent obesity in order to save societal resources. Moreover, international consensus is required on standardized methods to calculate the cost of obesity to improve homogeneity and comparability. This aspect should also be considered when including obesity-related diseases.

R. Williams

Francis Guillemin, C. Bombardier, D. Beaton

Tariq Alqahtani, H. Badreldin, Mohammed A. Alrashed et al.

Artificial Intelligence (AI) has revolutionized various domains, including education and research. Natural language processing (NLP) techniques and large language models (LLMs) such as GPT-4 and BARD have significantly advanced our comprehension and application of AI in these fields. This paper provides an in-depth introduction to AI, NLP, and LLMs, discussing their potential impact on education and research. By exploring the advantages, challenges, and innovative applications of these technologies, this review gives educators, researchers, students, and readers a comprehensive view of how AI could shape educational and research practices in the future, ultimately leading to improved outcomes. Key applications discussed in the field of research include text generation, data analysis and interpretation, literature review, formatting and editing, and peer review. AI applications in academics and education include educational support and constructive feedback, assessment, grading, tailored curricula, personalized career guidance, and mental health support. Addressing the challenges associated with these technologies, such as ethical concerns and algorithmic biases, is essential for maximizing their potential to improve education and research outcomes. Ultimately, the paper aims to contribute to the ongoing discussion about the role of AI in education and research and highlight its potential to lead to better outcomes for students, educators, and researchers.

Guendalina Caldarini, Sardar F. Jaf, K. McGarry

Chatbots are intelligent conversational computer systems designed to mimic human conversation to enable automated online guidance and support. The increased benefits of chatbots led to their wide adoption by many industries in order to provide virtual assistance to customers. Chatbots utilise methods and algorithms from two Artificial Intelligence domains: Natural Language Processing and Machine Learning. However, there are many challenges and limitations in their application. In this survey we review recent advances on chatbots, where Artificial Intelligence and Natural Language processing are used. We highlight the main challenges and limitations of current work and make recommendations for future research investigation

Lizhou Fan, Lingyao Li, Zihui Ma et al.

Large language models (LLMs), such as OpenAI's Generative Pre-trained Transformer (GPT), are a class of language models that have demonstrated outstanding performance across a range of natural language processing (NLP) tasks. LLMs have become a highly sought-after research area because of their ability to generate human-like language and their potential to revolutionize science and technology. In this study, we conduct bibliometric and discourse analyses of scholarly literature on LLMs. Synthesizing over 5,000 publications, this article serves as a roadmap for researchers, practitioners, and policymakers to navigate the current landscape of LLMs research. We present the research trends from 2017 to early 2023, identifying patterns in research paradigms and collaborations. We start with analyzing the core algorithm developments and NLP tasks that are fundamental in LLMs research. We then investigate the applications of LLMs in various fields and domains, including medicine, engineering, social science, and humanities. Our review also reveals the dynamic, fast-paced evolution of LLMs research. Overall, this article offers valuable insights into the current state, impact, and potential of LLMs research and its applications.

Rusmiyanto Rusmiyanto, Nining Huriati, Nining Fitriani et al.

In today's globalized world, the increasing need for English language ability has highlighted the necessity of good language acquisition and communication abilities. Artificial Intelligence (AI) has emerged as a viable aid in the field of education, including language acquisition, as technology advances. This study does a literature review to investigate the function of AI in the development of communication skills in English language learners. The goal of this research is to look at the existing research and literature on the use of AI-based technologies in English language learning environments. The essay opens with an overview of artificial intelligence and its possible uses in education. It then looks into the various methods in which AI might help English language learners strengthen their communication skills, including speaking, listening, reading, and writing. The findings of this literature review suggest that AI has the potential to significantly enhance English language learners' communication skills by providing personalized and interactive learning experiences. However, further research is needed to explore the long-term effects and optimal integration of AI in language learning environments. In conclusion, this article highlights the transformative role of AI in English language education and its potential to address the diverse needs of language learners. By understanding the current state of research and exploring the opportunities and challenges presented by AI in language learning, educators and policymakers can make informed decisions to harness the benefits of AI technology and maximize its impact on developing effective communication skills among English language learners.

Matheus Belarmino, Rackel Coelho, Roberto Lotudo et al.

Large Language Models (LLMs) have been increasingly used to optimize the analysis and synthesis of legal documents, enabling the automation of tasks such as summarization, classification, and retrieval of legal information. This study aims to conduct a systematic literature review to identify the state of the art in prompt engineering applied to LLMs in the legal context. The results indicate that models such as GPT-4, BERT, Llama 2, and Legal-Pegasus are widely employed in the legal field, and techniques such as Few-shot Learning, Zero-shot Learning, and Chain-of-Thought prompting have proven effective in improving the interpretation of legal texts. However, challenges such as biases in models and hallucinations still hinder their large-scale implementation. It is concluded that, despite the great potential of LLMs for the legal field, there is a need to improve prompt engineering strategies to ensure greater accuracy and reliability in the generated results.

Mingzhe Xing, Chang Tian, Jianan Zhang et al.

Modern large-scale networks introduce significant complexity in understanding network behaviors, increasing the risk of misconfiguration. Prior work proposed to understand network behaviors by mining network configurations, typically relying on domain-specific languages interfaced with formal models. While effective, they suffer from a steep learning curve and limited flexibility. In contrast, natural language (NL) offers a more accessible and interpretable interface, motivating recent research on NL-guided network behavior understanding. Recent advances in large language models (LLMs) further enhance this direction, leveraging their extensive prior knowledge of network concepts and strong reasoning capabilities. However, three key challenges remain: 1) numerous router devices with lengthy configuration files challenge LLM's long-context understanding ability; 2) heterogeneity across devices and protocols impedes scalability; and 3) complex network topologies and protocols demand advanced reasoning abilities beyond the current capabilities of LLMs. To tackle the above challenges, we propose NetMind, a novel framework for querying networks using NL. Our approach introduces a tree-based configuration chunking strategy to preserve semantic coherence while enabling efficient partitioning. We then construct a unified fact graph as an intermediate representation to normalize vendor-specific configurations. Finally, we design a hybrid imperative-declarative language to reduce the reasoning burden on LLMs and enhance precision. We contribute a benchmark consisting of NL question-answer pairs paired with network configurations. Experiments demonstrate that NetMind achieves accurate and scalable network behavior understanding, outperforming existing baselines.

Aleksandra Krasnodębska, Karolina Seweryn, Szymon Łukasik et al.

Despite increasing efforts to ensure the safety of large language models (LLMs), most existing safety assessments and moderation tools remain heavily biased toward English and other high-resource languages, leaving majority of global languages underexamined. To address this gap, we introduce a manually annotated benchmark dataset for language model safety classification in Polish. We also create adversarially perturbed variants of these samples designed to challenge model robustness. We conduct a series of experiments to evaluate LLM-based and classifier-based models of varying sizes and architectures. Specifically, we fine-tune three models: Llama-Guard-3-8B, a HerBERT-based classifier (a Polish BERT derivative), and PLLuM, a Polish-adapted Llama-8B model. We train these models using different combinations of annotated data and evaluate their performance, comparing it against publicly available guard models. Results demonstrate that the HerBERT-based classifier achieves the highest overall performance, particularly under adversarial conditions.

Zaenal Abidin, Akmal Junaidi, Wamiliana

Background: Stemming is significantly essential in natural language processing (NLP) due to the ability to minimize word variations to fundamental forms. This procedure facilitates the analysis of textual data and enhances the precision of classification and information retrieval. Objective: Previous related systematic literature review has not been conducted on stemming and lemmatization in regional languages in Indonesia. Therefore, this study aims to conduct a systematic literature review to capture the latest developments in stemming and lemmatization in regional languages in Indonesia. Methods: This study was carried out using Kitchenham method, analyzing 35 studies extracted from 740, which were obtained from Scopus, IEEE Xplore, and Google Scholar, and published between 2014 and 2023. Results: The results showed that study trends in stemming possessed the potential to continue developing every year. Additionally, the main element in stemming and lemmatization studies was found to be the availability of digital dictionaries in regional languages. This was because greater number of basic vocabularies contributed more positively to stemming or lemmatization. The availability of word morphology information in regional languages would be constructive for making rule-based stemmers. Meanwhile, corpus-based stemming and lemmatization studies could only be conducted for languages with a large corpus to ensure there were various affixed words to process. Conclusion: Based on SLR study, stemming and lemmatization in regional languages in Indonesia developed significantly from 2014 to 2023. The two main strategies applied included using available digital dictionaries and language morphology information. However, the main challenges encountered were the limited number of vocabulary words in the dictionaries and testing various rule-based methods.   Keywords: Lemmatization, Morphology, Rule-based, Stemming, Systematic Literature Review.

Artur Gałkowski, Rafał Zarębski

Debbie Loakes, Adele Gregory

(1) Background: Australian Aboriginal English (AAE) is a variety known to differ in various ways from the mainstream, but to date very little phonetic analysis has been carried out. This study is a description of L1 Aboriginal English in southern Australia, aiming to comprehensively describe the acoustics of vowels, focusing in particular on vowels known to be undergoing change in Mainstream Australian English. Previous work has focused on static measures of F1/F2, and here we expand on this by adding duration analyses, as well as dynamic F1/F2 measures. (2) Methods: This paper uses acoustic-phonetic analyses to describe the vowels produced by speakers of Aboriginal Australian English from two communities in southern Australia (Mildura and Warrnambool). The focus is vowels undergoing change in the mainstream variety–the short vowels in KIT, DRESS, TRAP, STRUT, LOT, and the long vowel GOOSE; focusing on duration, and static and dynamic F1/F2. As part of this description, we analyse the data using the sociophonetic variables gender, region, and age, and also compare the Aboriginal Australian English vowels to those of Mainstream Australian English. (3) Results: On the whole, for duration, few sociophonetic differences were observed. For static F1/F2, we saw that L1 Aboriginal English vowel spaces tend to be similar to Mainstream Australian English but can be analysed as more conservative (having undergone less change) as has also been observed for L2 Aboriginal English, in particular for KIT, DRESS, and TRAP. The Aboriginal English speakers had a less peripheral vowel space than Mainstream Australian English speakers. Dynamic analyses also highlighted dialectal differences between Aboriginal and Mainstream Australian English speakers, with greater F1/F2 movement in the trajectories of vowels overall for AAE speakers, which was more evident for some vowels (TRAP, STRUT, LOT, and GOOSE). Regional differences in vowel quality between the two locations were minimal, and more evident in the dynamic analyses. (4) Conclusions: This paper further highlights how Aboriginal Australian English is uniquely different from Mainstream Australian English with respect to certain vowel differences, and it also highlights some ways in which the varieties align. The differences, i.e., a more compressed vowel space, and greater F1/F2 movement in the trajectories of short vowels for AAE speakers, are specific ways that Aboriginal Australian English and Mainstream Australian English accents are different in these communities in the southern Australian state of Victoria.

Barış Can Aydın

This study aims to provide insights into understanding the theoretical background of the application of critical discourse analysis (CDA) in the translation of political texts in the field of translation studies. The study also casts light on the investigation into the ideological and discursive issues in translation through the use of CDA as well as political discourse and translation. CDA is crucial in understanding the role and significance of discourse in the translation of a political text without disregarding the literary sense, authentic style of the speaker in the target language, and rhetorical devices. In this regard, this study considers the case of a political speech to demonstrate the role and significance of CDA in the translation of political speech. For this reason, the study has selected the case of Donald Trump’s inaugural address for translation into the target language of Turkish by the study’s author through the use of a critical lens. Following a critical approach and Norman Fairclough’s (1995) model for CDA in the interpretation and translation of political discourse, this study aims to provide explanations and solutions to the difficulties encountered in the interpretation and translation of a political speech. Therefore, the comparison of the source text with the target text offered and discussed in this study helps to underline and raise awareness about the contributions of CDA to translation studies.

Maria Giulia Pinheiro, Saru Vidal, Cynthia Agra de Brito Neves

The purpose of this paper is to discuss and analyze public posts on the social network Facebook that point to xenophobic opinions about the (i)migrant poetry slam scene in Portugal. These posts reflect the post-colonial attitudes of a historically colonizing country, which is currently experiencing a cultural identity crisis that oscillates between Lusotropicalist and Eurocentric ideology, and whose economic, racial and gender discrimination is still present. We will therefore look at poetry slam in Portugal in the light of post-colonial, identity and social disputes.

Carl Edwards, Qingyun Wang, Lawrence Zhao et al.

Language-molecule models have emerged as an exciting direction for molecular discovery and understanding. However, training these models is challenging due to the scarcity of molecule-language pair datasets. At this point, datasets have been released which are 1) small and scraped from existing databases, 2) large but noisy and constructed by performing entity linking on the scientific literature, and 3) built by converting property prediction datasets to natural language using templates. In this document, we detail the $\textit{L+M-24}$ dataset, which has been created for the Language + Molecules Workshop shared task at ACL 2024. In particular, $\textit{L+M-24}$ is designed to focus on three key benefits of natural language in molecule design: compositionality, functionality, and abstraction.

Jafar Isbarov, Kavsar Huseynova, Elvin Mammadov et al.

The emergence of multilingual large language models has enabled the development of language understanding and generation systems in Azerbaijani. However, most of the production-grade systems rely on cloud solutions, such as GPT-4. While there have been several attempts to develop open foundation models for Azerbaijani, these works have not found their way into common use due to a lack of systemic benchmarking. This paper encompasses several lines of work that promote open-source foundation models for Azerbaijani. We introduce (1) a large text corpus for Azerbaijani, (2) a family of encoder-only language models trained on this dataset, (3) labeled datasets for evaluating these models, and (4) extensive evaluation that covers all major open-source models with Azerbaijani support.