Search Results (220)

This paper addresses text simplification task for Kazakh, a morphologically rich, low-resource language, by introducing KazSim, an instruction-tuned model built on multilingual large language models (LLMs). First, we develop a heuristic pipeline to identify complex Kazakh sentences, manually validating its performance on 400 examples and comparing it against a purely LLM-based selection method; we then use this pipeline to assemble a parallel corpus of 8709 complex–simple pairs via LLM augmentation. For the simplification task, we benchmark KazSim against standard Seq2Seq systems, domain-adapted Kazakh LLMs, and zero-shot instruction-following models. On an automatically constructed test set, KazSim (Llama-3.3-70B) achieves BLEU 33.50, SARI 56.38, and F1 87.56 with a length ratio of 0.98, outperforming all baselines. We also explore prompt language (English vs. Kazakh) and conduct human evaluation with three native speakers: KazSim scores 4.08 for fluency, 4.09 for meaning preservation, and 4.42 for simplicity—significantly above GPT-4o-mini. Error analysis shows that remaining failures cluster into tone change, tense change, and semantic drift, reflecting Kazakh’s agglutinative morphology and flexible syntax. Full article

Text2Cypher is a text-to-text task that converts natural language questions into Cypher queries. Recent research by Neo4j on Text2Cypher demonstrates that fine-tuning a baseline language model (a pretrained and instruction-tuned generative model) using a comprehensive Text2Cypher dataset can effectively enhance query generation performance. However, the improvement is still insufficient for effectively learning the syntax and semantics of complex natural texts, particularly when applied to unseen Cypher schema structures across diverse domains during training. To address this challenge, we propose a novel refinement training method based on baseline language models, employing reinforcement learning with Group Relative Policy Optimization (GRPO). This method leverages extracted semantic information, such as key-value properties and triple relationships from input texts during the training process. Experimental results of the proposed refinement training method applied to a small-scale baseline language model (SLM) like Qwen2.5-3B-Instruct demonstrate that it achieves competitive execution accuracy scores on unseen schemas across various domains. Furthermore, the proposed method significantly outperforms most baseline LMs with larger parameter sizes in terms of Google-BLEU and execution accuracy scores over Neo4j’s comprehensive Text2Cypher dataset, with the exception of colossal LLMs such as GPT4o, GPT4o-mini, and Gemini. Full article

Deep learning has enabled significant progress in source code generation, aiming to reduce the manual, error-prone, and time-consuming aspects of software development. While many existing models rely on recurrent neural networks (RNNs) with sequence-to-sequence architectures, these approaches struggle with the long and complex token sequences typical in source code. To address this, we propose a grammar-based convolutional neural network (CNN) combined with a tree-based representation to enhance accuracy and efficiency. Our model achieves state-of-the-art results on the benchmark HEARTHSTONE dataset, with a BLEU score of 81.4 and an Acc+ of 62.1%. We further evaluate the model on our proposed dataset, AST2CVCode, designed for computer vision applications, achieving 86.2 BLEU and 51.9% EM. Additionally, we introduce BLEU+, an enhanced evaluation metric tailored for functional correctness in code generation, which achieves a BLEU+ score of 92.0% on the AST2CVCode dataset. These results demonstrate the effectiveness of our approach in both model architecture and evaluation methodology. Full article

Telemedicine in emergency contexts presents unique challenges, particularly in multilingual and low-resource settings where accurate, clinical understanding and triage decision support are critical. This paper introduces TriagE-NLU, a novel multilingual natural language understanding system designed to perform both semantic parsing and clinical intervention classification from emergency dialogues. The system is built on a federated learning architecture to ensure data privacy and adaptability across regions and is trained using TriageX, a synthetic, clinically grounded dataset covering five languages (English, Spanish, Romanian, Arabic, and Mandarin). TriagE-NLU integrates fine-tuned multilingual transformers with a hybrid rules-and-policy decision engine, enabling it to parse structured medical information (symptoms, risk factors, temporal markers) and recommend appropriate interventions based on recognized patterns. Evaluation against strong multilingual baselines, including mT5, mBART, and XLM-RoBERTa, demonstrates superior performance by TriagE-NLU, achieving F1 scores of 0.91 for semantic parsing and 0.89 for intervention classification, along with 0.92 accuracy and a BLEU score of 0.87. These results validate the system’s robustness in multilingual emergency telehealth and its ability to generalize across diverse input scenarios. This paper establishes a new direction for privacy-preserving, AI-assisted triage systems. Full article

Background/Objectives: Extracting meaningful medical information from chest X-ray images and transcribing it into text is a complex task that requires a high level of expertise and directly affects clinical decision-making processes. Automatic reporting systems for this field in Turkish represent an important gap in scientific research, as they have not been sufficiently addressed in the existing literature. Methods: A deep learning-based approach called Model-SEY was developed with the aim of automatically generating Turkish medical reports from chest X-ray images. The Swin Transformer structure was used in the encoder part of the model to extract image features, while the text generation process was carried out using the cosmosGPT architecture, which was adapted specifically for the Turkish language. Results: With the permission of the ethics committee, a new dataset was created using image–report pairs obtained from Elazıg Fethi Sekin City Hospital and Indiana University Chest X-Ray dataset and experiments were conducted on this new dataset. In the tests conducted within the scope of the study, scores of 0.6412, 0.5335, 0.4395, 0.4395, 0.3716, and 0.2240 were obtained in BLEU-1, BLEU-2, BLEU-3, BLEU-4, and ROUGE word overlap evaluation metrics, respectively. Conclusions: Quantitative and qualitative analyses of medical reports autonomously generated by the proposed model have shown that they are meaningful and consistent. The proposed model is one of the first studies in the field of autonomous reporting using deep learning architectures specific to the Turkish language, representing an important step forward in this field. It will also reduce potential human errors during diagnosis by supporting doctors in their decision-making. Full article

Remote Sensing Image Change Captioning (RSICC) aims to generate natural language descriptions of changes in bi-temporal remote sensing images, providing more semantically interpretable results than conventional pixel-level change detection methods. However, existing approaches often rely on stacked Transformer modules, leading to suboptimal feature discrimination. Moreover, direct difference computation after feature extraction tends to retain task-irrelevant noise, limiting the model’s ability to capture meaningful changes. This study proposes a novel cross-spatial Transformer and symmetric difference localization network (CTSD-Net) for RSICC to address these limitations. The proposed Cross-Spatial Transformer adaptively enhances spatial-aware feature representations by guiding the model to focus on key regions across temporal images. Additionally, a hierarchical difference feature integration strategy is introduced to suppress noise by fusing multi-level differential features, while residual-connected high-level features serve as query vectors to facilitate bidirectional change representation learning. Finally, a causal Transformer decoder creates accurate descriptions by linking visual information with text. CTSD-Net achieved BLEU-4 scores of 66.32 and 73.84 on the LEVIR-CC and WHU-CDC datasets, respectively, outperforming existing methods in accurately locating change areas and describing them semantically. This study provides a promising solution for enhancing interpretability in remote sensing change analysis. Full article

Medical imaging is crucial for diagnosing, monitoring, and treating medical conditions. The medical reports of radiology images are the primary medium through which medical professionals can attest to their findings, but their writing is time-consuming and requires specialized clinical expertise. Therefore, the automated generation of radiography reports has the potential to improve and standardize patient care and significantly reduce the workload of clinicians. Through our work, we have designed and evaluated an end-to-end transformer-based method to generate accurate and factually complete radiology reports for X-ray images. Additionally, we are the first to introduce curriculum learning for end-to-end transformers in medical imaging and demonstrate its impact in obtaining improved performance. The experiments were conducted using the MIMIC-CXR-JPG database, the largest available chest X-ray dataset. The results obtained are comparable with the current state of the art on the natural language generation (NLG) metrics BLEU and ROUGE-L, while setting new state-of-the-art results on F1 examples-averaged F1-macro and F1-micro metrics for clinical accuracy and on the METEOR metric widely used for NLG. Full article

This article introduces a freely available Spanish–Mixtec parallel corpus designed to foster natural language processing (NLP) development for an indigenous language that remains digitally low-resourced. The dataset, comprising 14,587 sentence pairs, covers Mixtec variants from Guerrero (Tlacoachistlahuaca, Northern Guerrero, and Xochapa) and Oaxaca (Western Coast, Southern Lowland, Santa María Yosoyúa, Central, Lower Cañada, Western Central, San Antonio Huitepec, Upper Western, and Southwestern Central). Texts are classified into four main domains as follows: education, law, health, and religion. To compile these data, we conducted a two-phase collection process as follows: first, an online search of government portals, religious organizations, and Mixtec language blogs; and second, an on-site retrieval of physical texts from the library of the Autonomous University of Querétaro. Scanning and optical character recognition were then performed to digitize physical materials, followed by manual correction to fix character misreadings and remove duplicates or irrelevant segments. We conducted a preliminary evaluation of the collected data to validate its usability in automatic translation systems. From Spanish to Mixtec, a fine-tuned GPT-4o-mini model yielded a BLEU score of 0.22 and a TER score of 122.86, while two fine-tuned open source models mBART-50 and M2M-100 yielded BLEU scores of 4.2 and 2.63 and TER scores of 98.99 and 104.87, respectively. All code demonstrating data usage, along with the final corpus itself, is publicly accessible via GitHub and Figshare. We anticipate that this resource will enable further research into machine translation, speech recognition, and other NLP applications while contributing to the broader goal of preserving and revitalizing the Mixtec language. Full article

The environmental perception capability of intelligent ships is essential for enhancing maritime navigation safety and advancing shipping intelligence. Image caption generation technology plays a pivotal role in this context by converting visual information into structured semantic descriptions. However, existing general purpose models often struggle to perform effectively in complex maritime environments due to limitations in visual feature extraction and semantic modeling. To address these challenges, this study proposes a transformer dual-stream information (TDSI) model. The proposed model uses a Swin-transformer to extract grid features and combines them with fine-grained scene semantics obtained via SegFormer. A dual-encoder structure independently encodes the grid and segmentation features, which are subsequently fused through a feature fusion module for implicit integration. A decoder with a cross-attention mechanism is then employed to generate descriptive captions for maritime images. Extensive experiments were conducted using the constructed maritime semantic segmentation and maritime image captioning datasets. The results demonstrate that the proposed TDSI model outperforms existing mainstream methods in terms of several evaluation metrics, including BLEU, METEOR, ROUGE, and CIDEr. These findings confirm the effectiveness of the TDSI model in enhancing image captioning performance in maritime environments. Full article

With the development of artificial intelligence and intelligent sensor technologies, traditional legal English teaching approaches have faced numerous challenges in handling multimodal inputs and complex reasoning tasks. In response to these challenges, a cross-modal legal English question-answering system based on visual and acoustic sensor inputs was proposed, integrating image, text, and speech information and adopting a unified vision–language–speech encoding mechanism coupled with dynamic attention modeling to effectively enhance learners’ understanding and expressive abilities in legal contexts. The system exhibited superior performance across multiple experimental evaluations. In the assessment of question-answering accuracy, the proposed method achieved the best results across BLEU, ROUGE, Precision, Recall, and Accuracy, with an Accuracy of 0.87, Precision of 0.88, and Recall of 0.85, clearly outperforming the traditional ASR+SVM classifier, image-retrieval-based QA model, and unimodal BERT QA system. In the analysis of multimodal matching performance, the proposed method achieved optimal results in Matching Accuracy, Recall@1, Recall@5, and MRR, with a Matching Accuracy of 0.85, surpassing mainstream cross-modal models such as VisualBERT, LXMERT, and CLIP. The user study further verified the system’s practical effectiveness in real teaching environments, with learners’ understanding improvement reaching 0.78, expression improvement reaching 0.75, and satisfaction score reaching 0.88, significantly outperforming traditional teaching methods and unimodal systems. The experimental results fully demonstrate that the proposed cross-modal legal English question-answering system not only exhibits significant advantages in multimodal feature alignment and deep reasoning modeling but also shows substantial potential in enhancing learners’ comprehensive capabilities and learning experiences. Full article

Synthetic medical text generation has emerged as a solution to data scarcity and privacy constraints in clinical NLP. This review systematically evaluates the use of Large Language Models (LLMs) for structured medical text generation, examining techniques such as retrieval-augmented generation (RAG), structured fine-tuning, and domain-specific adaptation. Four search queries were applied following the PRISMA methodology to identify and extract data from 153 studies. Key benchmarking metrics, such as performance measures, and qualitative insights, including methodological trends and challenges, were documented. The results show that while LLM-generated text improves fluency, hallucinations and factual inconsistencies persist. Structured consultation models, such as SOAP and Calgary–Cambridge, enhance coherence but do not fully prevent errors. Hybrid techniques that combine retrieval-based grounding with domain-specific fine-tuning improve factual accuracy and task performance. Conventional evaluation metrics (e.g., ROUGE, BLEU) are insufficient for medical validation, highlighting the need for domain-specific benchmarks. Privacy-preserving strategies, including differential privacy and PHI de-identification, support regulatory compliance but may reduce linguistic quality. These findings are relevant for clinical NLP applications, such as AI-powered scribe systems, where structured synthetic datasets can improve transcription accuracy and documentation reliability. The conclusions highlight the need for balanced approaches that integrate medical structure, factual control, and privacy to enhance the usability of synthetic medical text. Full article

This study investigates translation quality between Arabic and English, comparing traditional rule-based machine translation systems, modern neural machine translation tools such as Google Translate, and large language models like ChatGPT. The research adopts both qualitative and quantitative approaches to assess the efficacy, accuracy, and contextual fidelity of translations. It particularly focuses on the translation of idiomatic and colloquial expressions as well as technical texts and genres. Using well-established evaluation metrics such as bilingual evaluation understudy (BLEU), translation error rate (TER), and character n-gram F-score (chrF), alongside the qualitative translation quality assessment model proposed by Juliane House, this study investigates the linguistic and semantic nuances of translations generated by different systems. This study concludes that although metric-based evaluations like BLEU and TER are useful, they often fail to fully capture the semantic and contextual accuracy of idiomatic and expressive translations. Large language models, particularly ChatGPT, show promise in addressing this gap by offering more coherent and culturally aligned translations. However, both systems demonstrate limitations that necessitate human post-editing for high-stakes content. The findings support a hybrid approach, combining machine translation tools with human oversight for optimal translation quality, especially in languages with complex morphology and culturally embedded expressions like Arabic. Full article

Accurate classification of poultry behavior is critical for assessing welfare and health, yet most existing methods predict behavior categories without providing explanations for the image content. This study introduces the PBC-Transformer model, a novel model that integrates image captioning techniques to enhance poultry behavior classification, mimicking expert assessment processes. The model employs a multi-head concentrated attention mechanism, Head Spatial Position Coding (HSPC), to enhance spatial information; a learnable sparse mechanism (LSM) and RNorm function to reduce noise and strengthen feature correlation; and a depth-wise separable convolutional network for improved local feature extraction. Furthermore, a multi-level attention differentiator dynamically selects image regions for precise behavior descriptions. To balance caption generation with classification, we introduce the ICL-Loss function, which adaptively adjusts loss weights. Extensive experiments on the PBC-CapLabels dataset demonstrate that PBC-Transformer outperforms 13 commonly used classification models, improving accuracy by 15% and achieving the highest scores across image captioning metrics: Bleu4 (0.498), RougeL (0.794), Meteor (0.393), and Spice (0.613). Full article

In recent decades, sustainability and biophilic design have gained significant attention as revived concepts in architecture, offering innovative pathways to reconnect the built environment with nature. Can these principles be characterized and assessed in vernacular architectural contexts so as to be incorporated into contemporary sustainable practices? This research seeks to answer this question by examining the vernacular architecture of Kasbahs and Ksour in southern Morocco through the lens of biophilic design. The link between the two remains underexplored, specifically in the context of southern Morocco—a gap this article seeks to address. This research analyzes these heritage architectures by combining a theoretical exploration of sustainability, biophilic design (BD), and operational BD frameworks with a practical evaluation using a Biophilic Interior Design Matrix. This analysis is particularly pertinent as the contemporary society spends roughly 90% of its time indoors and is considered to be an “indoor generation”. After examining eleven vernacular buildings spread over key areas of Ouarzazate Province in southern Morocco against 54 biophilic design attributes, the findings reveal that Kasbahs and Ksour showcase sustainability and biophilic qualities. This demonstrates that Moroccan traditional architectural values can enable heritage preservation through biophilic principles to deliver culturally contextual and sustainable architectural solutions for contemporary practice. Full article

Code-to-code translation, a critical domain in software engineering, increasingly utilizes trans-compilers to translate between high-level languages. Traditionally, the fidelity of such translations has been evaluated using the BLEU score, which predominantly measures token similarity between the generated output and the ground truth. However, this metric falls short of assessing the methodologies underlying the translation processes and only evaluates the translations that are tested. To bridge this gap, this paper introduces an innovative architecture, “TC-Verifier”, to formally employ the Uppaal Model-checker to verify trans-compiler-based code translators. We applied the proposed architecture to a trans-compiler translating between Swift and Java, providing insights into the verified and unverified aspects of the translation process. Our findings illuminate the strengths and limitations of using Model-checking for formal verification in code translation. Notably, the examined trans-compiler reached a verification success rate of 50.74% for the grammar rules and productions modeled. This study underscores the gaps in trans-compiler-based translations and suggests that these gaps could potentially be addressed by integrating Large Language Models (LLMs) in future work. Full article