Hasil untuk "Computational linguistics. Natural language processing"

Neha Tyagi, Deepshikha Bhargava, Anil Ahlawat

Abstract Swarm Intelligence (SI) has become a strong paradigm for numerical optimization, which has inspired a wide range of metaheuristic algorithms. This paper presents the Hybrid Mating Optimization (HMO), a novel bio-inspired algorithm that synergically merges four mating and communication behaviors of nature: butterfly pheromone navigation (global exploration), honeybee foraging (local exploitation), red deer dominance selection (adaptive hierarchy), and woodpecker rhythmic perturbation (diversity preservation). This hybrid mechanism is able to find a dynamic balance between exploration and exploitation, without causing premature convergence. Extensive experiments on the CEC-2017 benchmark suite show that HMO can converge faster and obtain higher accuracy than state-of-the-art algorithms such as PSO, DE, EHO, and CMA-ES. The statistical significance is further verified using Wilcoxon signed-rank tests and t-tests. HMO also has scalability both in unimodal and multimodal environments. Furthermore, a real-world case study of an engineering problem on pressure vessel design validates the effectiveness of HMO in constrained optimization problems.

Samia BERBRA & Asma MARIR

Résumé : Ciblant un public en plein éducation, la littéraire de jeunesse a toujours eu comme devise : éduquer en s’amusant. Dans cet article, nous vérifions le principe de ce genre en nous appuyant sur le best seller italien Le fabricant de larmes d’Erin Doom pour savoir en quoi cette œuvre s’inscrit-elle dans la littérature de jeunesse tout en touchant à la fois des aspects divertissants et philosophiques ?  Étant une œuvre destinée beaucoup plus aux adolescents, voire aux jeunes adultes, nous nous interrogeons plus précisément sur les messages adressés à ces jeunes lecteurs concernant les épreuves et les défis de la vie. Ainsi, notre démarche se propose, dans un premier temps, comme une lecture thématique pour dégager les éléments caractéristiques de la littérature de jeunesse présents dans l’œuvre-corpus et, dans un second temps, souligner les différents messages éducatifs, voire philosophiques. Mots-clés : Littérature de jeunesse ; Jeunes adultes ; Divertissement ; Philosophie.

Kate Takyi, Rose-Mary Owusuaa Mensah Gyening, Shester Msouobu Gueuwou et al.

Abstract Research on sign language translation is ongoing with a high social inclusive goal of crossing the bridge between people with hearing disability using sign language as their basic way to communicate to others who do not understand sign language. Hundreds of different sign languages exist instead of a single universal sign language. Research on translating sign languages from high-income nations has grown significantly, but little is known about translating sign languages from Africa. In this paper, we curate a novel video-to-text African sign languages translation dataset containing sign language videos of Bible verses from six (6) different African countries. We experimented with competitive machine translation and sign language translation techniques on our dataset, including the application of transformers to sign language translation, multilingual training, and cross-transfer learning. We evaluated them in terms of accuracy and precision. The results from our experiments prove that having one Multilingual model for all the languages tends to be a better choice when deployed in real system in terms of memory usage with an accuracy of 94.6% and precision of 97.3%. These results give headway for more multilingual models to be developed to enhance inclusion for the deaf community and bridge the gap between the hearing and the deaf in Africa.

Fengqiang Gao, Qingyuan Zhu, Guifang Shao et al.

Abstract Efficient detection of surface defects is primary for ensuring product quality during manufacturing processes. To enhance the performance of deep learning‐based methods in practical applications, the authors propose Dense‐YOLO, a fast surface defect detection network that combines the strengths of DenseNet and you only look once version 3 (YOLOv3). The authors design a lightweight backbone network with improved densely connected blocks, optimising the utilisation of shallow features while maintaining high detection speeds. Additionally, the authors refine the feature pyramid network of YOLOv3 to increase the recall of tiny defects and overall positioning accuracy. Furthermore, an online multi‐angle template matching technique is introduced based on normalised cross‐correlation to precisely locate the detection area. This refined template matching method not only accelerates detection speed but also mitigates the influence of the background. To validate the effectiveness of our enhancements, the authors conduct comparative experiments across two private datasets and one public dataset. Results show that Dense‐YOLO outperforms existing methods, such as faster R‐CNN, YOLOv3, YOLOv5s, YOLOv7, and SSD, in terms of mean average precision (mAP) and detection speed. Moreover, Dense‐YOLO outperforms networks inherited from VGG and ResNet, including improved faster R‐CNN, FCOS, M2Det‐320 and FRCN, in mAP.

Pa. Andal, M. Thangaraj

Abstract Automatic image classification for plant leaf disease identification (LDI) is an important task in computer vision, food processing, robotics and precision agriculture. Humans classify diseases that occur in plants by careful examination of leaves via visual inspection and knowledge of diseases that can occur to various plants. Similarly, digital images of leaves acquired through devices or from existing datasets are examined via computer vision techniques and classified with trained knowledge using various artificial intelligence approaches. This review discusses several methods explored by various researchers and visualizes their results in terms of the accuracies achieved and also the limitations of each method related to leaf disease identification. The objective of this article is to present a comprehensive review of recent research works by briefly describing the nature, size of data, No of plants and diseases covered, steps involved in the classification approaches, performance and limitation. Thus this article gives overview of recent machine learning (ML), deep learning (DL) and other approaches in LDI. The correctness of LDI classification is presented to the readers by mentioning the accuracy as found from the research articles. Apart from this other efficiency considerations are presented as when needed for describing the research work. This would give a quick summary of review of latest LDI research works by providing black box presentation to the approaches without elaborating the detail steps of the chosen approach.

John Pavlopoulos, Juli Bakagianni, Kanella Pouli et al.

Natural Language Processing (NLP) for lesser-resourced languages faces persistent challenges, including limited datasets, inherited biases from high-resource languages, and the need for domain-specific solutions. This study addresses these gaps for Modern Greek through three key contributions. First, we evaluate the performance of open-source (Llama-70b) and closed-source (GPT-4o mini) large language models (LLMs) on seven core NLP tasks with dataset availability, revealing task-specific strengths, weaknesses, and parity in their performance. Second, we expand the scope of Greek NLP by reframing Authorship Attribution as a tool to assess potential data usage by LLMs in pre-training, with high 0-shot accuracy suggesting ethical implications for data provenance. Third, we showcase a legal NLP case study, where a Summarize, Translate, and Embed (STE) methodology outperforms the traditional TF-IDF approach for clustering \emph{long} legal texts. Together, these contributions provide a roadmap to advance NLP in lesser-resourced languages, bridging gaps in model evaluation, task innovation, and real-world impact.

Jurek Eisinger, Ward Gauderis, Lin de Huybrecht et al.

The Categorical Compositional Distributional (DisCoCat) framework models meaning in natural language using the mathematical framework of quantum theory, expressed as formal diagrams. DisCoCat diagrams can be associated with tensor networks and quantum circuits. DisCoCat diagrams have been connected to density matrices in various contexts in Quantum Natural Language Processing (QNLP). Previous use of density matrices in QNLP entails modelling ambiguous words as probability distributions over more basic words (the word \texttt{queen}, e.g., might mean the reigning queen or the chess piece). In this article, we investigate using probability distributions over processes to account for syntactic ambiguity in sentences. The meanings of these sentences are represented by density matrices. We show how to create probability distributions on quantum circuits that represent the meanings of sentences and explain how this approach generalises tasks from the literature. We conduct an experiment to validate the proposed theory.

Jian Zhang, Jia Shao

While large language models (LLMs) have introduced novel paradigms in science and education, their adoption in higher education is constrained by inherent limitations. These include a tendency to produce inaccuracies and high computational requirements, which compromise the strict demands for accurate and reliable knowledge essential in higher education. Small language models (MiniLMs), by contrast, offer distinct advantages in professional education due to their lightweight nature and precise retrieval capabilities. This research takes "Atmospheric Physics" as an example. We established a specialized corpus and image repository by gathering over 550,000 full-text PDFs from over 130 international well-respected journals in Earth and environmental science. From this collection, we extracted over 100 million high-quality sentence-level corpus and more than 3 million high-resolution academic images. Using MiniLMs, these resources were organized into a high-dimensional vector library for precise retrieval and efficient utilization of extensive educational content. Consequently, we systematically redesigned the courses, textbooks, and teaching strategies for "Atmospheric Physics" based on MiniLMs. The course is designed as a "interdisciplinary-frontier" system, breaking down traditional boundaries between atmospheric science, space science, hydrology, and remote sensing. Teaching materials are transformed from static, lagging text formats into a dynamic digital resource library powered by MiniLM. For teaching methods, we have designed a question-based learning pathway. This paradigm promotes a shift from passive knowledge transfer to active cognitive development. Consequently, this MiniLM-driven "Atmospheric Physics" course demonstrates a specific avenue for "AI for education".

Chenqian Le, Ziheng Gong, Chihang Wang et al.

Large language models (LLMs) have shown great potential in medical question answering (MedQA), yet adapting them to biomedical reasoning remains challenging due to domain-specific complexity and limited supervision. In this work, we study how prompt design and lightweight fine-tuning affect the performance of open-source LLMs on PubMedQA, a benchmark for multiple-choice biomedical questions. We focus on two widely used prompting strategies - standard instruction prompts and Chain-of-Thought (CoT) prompts - and apply QLoRA for parameter-efficient instruction tuning. Across multiple model families and sizes, our experiments show that CoT prompting alone can improve reasoning in zero-shot settings, while instruction tuning significantly boosts accuracy. However, fine-tuning on CoT prompts does not universally enhance performance and may even degrade it for certain larger models. These findings suggest that reasoning-aware prompts are useful, but their benefits are model- and scale-dependent. Our study offers practical insights into combining prompt engineering with efficient finetuning for medical QA applications.

Judith Stelter

The COVID-19 pandemic has caused uncertainty in society and thus created a breeding ground for fake news. In social media in particular, targeted disinformation was and is practiced in order to emotionalize, confuse and manipulate users to achieve political and economic goals. In keeping with its educational and orienting function and in order to curb the spread of fake news, the established news magazine DER SPIEGEL carries out so-called fact checks and publishes the results on the social network Instagram. Although the text type “fact check” is considered young and poorly defined, its importance in socially uncertain times and its potential for a quick, safe and educational function are clear. In order to approach the text type “fact check” specifically in a social media environment, a selected corpus of 12 Instagram posts from @spiegelmagazin in the period between March and May 2020 will be examined using a multimodal, qualitative discourse analysis. Taking an inductive perspective, we ask which strategic text-image strategies can be found in the Instagram posts, which communicative functions are fulfilled and which overarching intentions should be taken into account. The study itself is intended to offer a first approach to the text type “fact check” and open the view to further, relevant questions.

Imane KERROUD, Lamia DEBBECHE & Lydia BOUCHAMA

Abstract: Algeria is witnessing rapid rural-urban transformations, especially after independence, with an increase in cities and their populations, leading to significant social and economic changes. This study aims to analyze the rural-urban transformations in the F'Kirina region in the Oum El Bouaghi Wilaya, based on a methodology that included literature review, map analysis, and field visits. The results show that F'Kirina is characterized by a strong agricultural sector and an active livestock sector, despite limited industrial activity. The trade and services sector has grown due to new economic policies. These changes highlight the region's shift towards an urban character, reflecting Algeria’s accelerating urban dynamics. Keywords: Algeria, F'Kirina, Economic Development, Rural-Urban Urbanization, Urban Expansion

Nisrine SAHNOUNE

Abstract: This study seeks to examine the language attitudes of Algerian students regarding the use of French as the language of instruction in scientific fields within Algeria. It aims to understand students' perspectives on French as a language of instruction in Medical sciences. It also investigates how language proficiency —and the pervasive use and growing dominance of French in these areas—influences students’ language use and attitudes. To achieve the objectives of this research, a questionnaire was administered to a sample of 100 participants. The findings show a prevalent use of Algerian Arabic (AA) compared to French and Modern Standard Arabic (MSA). The results also reveal that all participants demonstrated positive attitudes toward MSA while expressing negative attitudes toward French. The study also reveals some unexpected results, showing a strong preference for AA as a potential language of instruction for engaging with scientific knowledge and participating in scientific discourse. This highlights the urgent need to consider transitioning toward an English-based educational system, particularly in scientific disciplines. Furthermore, the findings suggest that despite the participants' high proficiency in using MSA as a medium of instruction, it is not the preferred language within the educational system for scientific fields. Ultimately, this context highlights the significant role that proficiency level plays in shaping students’ language preferences. Keywords: proficiency level, language use, attitudes, Modern Standard Arabic (MSA), French, Algerian Arabic, language of instruction.

Pascal Sikorski, Kaleb Yu, Lucy Billadeau et al.

This paper introduces a prototype for a new approach to assistive robotics, integrating edge computing with Natural Language Processing (NLP) and computer vision to enhance the interaction between humans and robotic systems. Our proof of concept demonstrates the feasibility of using large language models (LLMs) and vision systems in tandem for interpreting and executing complex commands conveyed through natural language. This integration aims to improve the intuitiveness and accessibility of assistive robotic systems, making them more adaptable to the nuanced needs of users with disabilities. By leveraging the capabilities of edge computing, our system has the potential to minimize latency and support offline capability, enhancing the autonomy and responsiveness of assistive robots. Experimental results from our implementation on a robotic arm show promising outcomes in terms of accurate intent interpretation and object manipulation based on verbal commands. This research lays the groundwork for future developments in assistive robotics, focusing on creating highly responsive, user-centric systems that can significantly improve the quality of life for individuals with disabilities. For video demonstrations and source code, please refer to: https://tinyurl.com/EnhancedArmEdgeNLP.

Jiaxuan Li, Richard Futrell

The functional interpretation of language-related ERP components has been a central debate in psycholinguistics for decades. We advance an information-theoretic model of human language processing in the brain in which incoming linguistic input is processed at first shallowly and later with more depth, with these two kinds of information processing corresponding to distinct electroencephalographic signatures. Formally, we show that the information content (surprisal) of a word in context can be decomposed into two quantities: (A) shallow surprisal, which signals shallow processing difficulty for a word, and corresponds with the N400 signal; and (B) deep surprisal, which reflects the discrepancy between shallow and deep representations, and corresponds to the P600 signal and other late positivities. Both of these quantities can be estimated straightforwardly using modern NLP models. We validate our theory by successfully simulating ERP patterns elicited by a variety of linguistic manipulations in previously-reported experimental data from six experiments, with successful novel qualitative and quantitative predictions. Our theory is compatible with traditional cognitive theories assuming a `good-enough' shallow representation stage, but with a precise information-theoretic formulation. The model provides an information-theoretic model of ERP components grounded on cognitive processes, and brings us closer to a fully-specified neuro-computational model of language processing.

Farid Ariai, Joel Mackenzie, Gianluca Demartini

Natural Language Processing (NLP) is revolutionising the way both professionals and laypersons operate in the legal field. The considerable potential for NLP in the legal sector, especially in developing computational assistance tools for various legal processes, has captured the interest of researchers for years. This survey follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses framework, reviewing 154 studies, with a final selection of 131 after manual filtering. It explores foundational concepts related to NLP in the legal domain, illustrating the unique aspects and challenges of processing legal texts, such as extensive document lengths, complex language, and limited open legal datasets. We provide an overview of NLP tasks specific to legal text, such as Document Summarisation, Named Entity Recognition, Question Answering, Argument Mining, Text Classification, and Judgement Prediction. Furthermore, we analyse both developed legal-oriented language models, and approaches for adapting general-purpose language models to the legal domain. Additionally, we identify sixteen open research challenges, including the detection and mitigation of bias in artificial intelligence applications, the need for more robust and interpretable models, and improving explainability to handle the complexities of legal language and reasoning.

Moreno La Quatra, Alkis Koudounas, Elena Baralis et al.

Italy exhibits rich linguistic diversity across its territory due to the distinct regional languages spoken in different areas. Recent advances in self-supervised learning provide new opportunities to analyze Italy's linguistic varieties using speech data alone. This includes the potential to leverage representations learned from large amounts of data to better examine nuances between closely related linguistic varieties. In this study, we focus on automatically identifying the geographic region of origin of speech samples drawn from Italy's diverse language varieties. We leverage self-supervised learning models to tackle this task and analyze differences and similarities between Italy's regional languages. In doing so, we also seek to uncover new insights into the relationships among these diverse yet closely related varieties, which may help linguists understand their interconnected evolution and regional development over time and space. To improve the discriminative ability of learned representations, we evaluate several supervised contrastive learning objectives, both as pre-training steps and additional fine-tuning objectives. Experimental evidence shows that pre-trained self-supervised models can effectively identify regions from speech recording. Additionally, incorporating contrastive objectives during fine-tuning improves classification accuracy and yields embeddings that distinctly separate regional varieties, demonstrating the value of combining self-supervised pre-training and contrastive learning for this task.

Abadlia NASSIMA

Résumé : Cet article s’interroge sur les questions d’« l’interculturel » et de « transnational » dans les littératures franco-africaines, tels que représentés dans le concept « d’afroplitanisme » que l’on doit à Achille Mbembé. Inscrit dans un contexte de revendication nationaliste, d’un passé/présent colonial/post-colonial, d’une culture diasporique africaine, mais aussi d’un contexte de mondialisation et projet de mobilisation internationale, voire mondiale. En quoi ce concept « d’afropolitanisme », tel que représenté dans le roman Voici venir les rêveurs d’Imbolo Mbue définit-il l’identité africaine, et les dimensions d’une culture diasporique, de solidarité nationale, raciale ou transnationale des diasporas noires dans le monde, en marche avec les progrès de la mondialisation ? Mots clés : Afropolitanisme-transnational-interculturel-diaspora-identité.

Muhammad Liaqat, Dr. Altaf Yousafzai

Munshi Prem Chand is a prominent name in Urdu novel. He successfully put various subjects in his novels. He is a native of a country where different cultures were born. Common Indian culture is a golden chapter in history. Prem Chand deeply observed Islamic culture and Islamic teachings and concluded that Islam and Islamic teachings cannot be compared to any culture or religion in the world. He thinks that no religion is bad, but the bad behavior of its followers makes it worse. The same qualities of his personality are also in his writings. He studied islam deeply and called its teachings extremely useful to humanity. In this article the author has elaborated that Islamic society and its values are unprecedented and cannot be compared to any religion in the world. <div><br /></div>

Aida Mostafazadeh Davani, Mohammad Atari, Brendan Kennedy et al.

AbstractSocial stereotypes negatively impact individuals’ judgments about different groups and may have a critical role in understanding language directed toward marginalized groups. Here, we assess the role of social stereotypes in the automated detection of hate speech in the English language by examining the impact of social stereotypes on annotation behaviors, annotated datasets, and hate speech classifiers. Specifically, we first investigate the impact of novice annotators’ stereotypes on their hate-speech-annotation behavior. Then, we examine the effect of normative stereotypes in language on the aggregated annotators’ judgments in a large annotated corpus. Finally, we demonstrate how normative stereotypes embedded in language resources are associated with systematic prediction errors in a hate-speech classifier. The results demonstrate that hate-speech classifiers reflect social stereotypes against marginalized groups, which can perpetuate social inequalities when propagated at scale. This framework, combining social-psychological and computational-linguistic methods, provides insights into sources of bias in hate-speech moderation, informing ongoing debates regarding machine learning fairness.

Wenyue Hua, Lifeng Jin, Linfeng Song et al.

Abstract Pretrained natural language processing (NLP) models have achieved high overall performance, but they still make systematic errors. Instead of manual error analysis, research on slice detection models (SDMs), which automatically identify underperforming groups of datapoints, has caught escalated attention in Computer Vision for both understanding model behaviors and providing insights for future model training and designing. However, little research on SDMs and quantitative evaluation of their effectiveness have been conducted on NLP tasks. Our paper fills the gap by proposing a benchmark named “Discover, Explain, Improve (DEIm)” for classification NLP tasks along with a new SDM Edisa. Edisa discovers coherent and underperforming groups of datapoints; DEIm then unites them under human-understandable concepts and provides comprehensive evaluation tasks and corresponding quantitative metrics. The evaluation in DEIm shows that Edisa can accurately select error-prone datapoints with informative semantic features that summarize error patterns. Detecting difficult datapoints directly boosts model performance without tuning any original model parameters, showing that discovered slices are actionable for users.1