Search Results (829)

Large language models and retrieval-augmented generation (RAG) are widely applied in knowledge question-answering tasks. However, in knowledge-intensive domains such as agriculture, hallucination and insufficient retrieval accuracy remain challenging. To address these issues, we propose Sem-RAG, a corn planting knowledge question-answering algorithm based on fine-grained semantic retrieval enhancement. Unlike standard NaiveRAG, which retrieves only fixed-length text chunks, and GraphRAG, which relies solely on graph node connections, Sem-RAG introduces a dual-store retrieval mechanism. It constructs both a surface semantic store (chunk-level embeddings) and a fine-grained semantic store derived from Leiden-based community summaries. These community summaries do not merely shorten contexts; instead, they provide thematic-level semantic aggregation across document chunks, thereby enhancing semantic coverage and reducing noise. During retrieval, user queries are matched against the surface store to locate relevant chunks and simultaneously linked to corresponding thematic summaries in the fine-grained store, ensuring that both local details and higher-level associations are leveraged. We evaluated Sem-RAG on the corn knowledge question-answering dataset CornData. The algorithm achieved Answer-C, Answer-R, and CR scores of 94.6%, 84.6%, and 70.4%, respectively, which were 2.6%, 1.7%, and 1.6% higher than those of traditional NaiveRAG. These results demonstrate that Sem-RAG materially improves the quality and reliability of agricultural knowledge Q&A by combining dual-store retrieval with community-level semantic aggregation. Full article

This paper introduces a hypergraph-based formalism for modeling kinematic and dynamic structures in robotics, addressing limitations of the existing formats such as Unified Robot Description Format (URDF), MuJoCo-XML, and Simulation Description Format (SDF). Our method represents mechanical constraints and connections as hyperedges, enabling the native description of multi-joint closures, tendon-driven actuation, and multi-physics coupling. We present a tensor-based representation derived via star-expansion, implemented in the Hypergraph Model Cognition Framework (HyMeKo) language. Comparative experiments show a substantial reduction in model verbosity compared to URDF while retaining expressiveness for large-language model integration. The approach is demonstrated on simple robotic arms and a quarter vehicle model, with derived state-space equations. This work suggests that hypergraph-based models can provide a modular, compact, and semantically rich alternative for the next-generation simulation and design workflows. The introduced formalism reaches 50% reduction compared to URDF descriptions and 20% reduction compared to MuJoCo-XML descriptions. Full article

The accurate prediction of herb–target interactions is essential for the modernization of traditional Chinese medicine (TCM) and the advancement of drug discovery. Nonetheless, the inherent complexity of herbal compositions and diversity of molecular targets render experimental validation both time-consuming and labor-intensive. We propose a graph neural network model, MAMGN-HTI, which integrates metapaths with attention mechanisms. A heterogeneous graph consisting of herbs, efficacies, ingredients, and targets is constructed, where semantic metapaths capture latent relationships among nodes. An attention mechanism is employed to dynamically assign weights, thereby emphasizing the most informative metapaths. In addition, ResGCN and DenseGCN architectures are combined with cross-layer skip connections to improve feature propagation and enable effective feature reuse. Experiments show that MAMGN-HTI outperforms several state-of-the-art methods across multiple metrics, exhibiting superior accuracy, robustness, and generalizability in HTI prediction and candidate drug screening. Validation against literature and databases further confirms the model’s predictive reliability. The model also successfully identified herbs with potential therapeutic effects for hyperthyroidism, including Vinegar-processed Bupleuri Radix (Cu Chaihu), Prunellae Spica (Xiakucao), and Processed Cyperi Rhizoma (Zhi Xiangfu). MAMGN-HTI provides a reliable computational framework and theoretical foundation for applying TCM in hyperthyroidism treatment, providing mechanistic insights while improving research efficiency and resource utilization. Full article

Manual pixel-level annotation remains a major bottleneck in deploying deep learning models for dense prediction and semantic segmentation tasks across domains. This challenge is especially pronounced in applications involving fine-scale structures, such as cracks in infrastructure or lesions in medical imaging, where annotations are time-consuming, expensive, and subject to inter-observer variability. To address these challenges, this work proposes a weakly supervised and annotation-efficient segmentation framework that integrates sparse bounding-box annotations with a limited subset of strong (pixel-level) labels to train robust segmentation models. The fundamental element of the framework is a lightweight Bounding Box Encoder that converts weak annotations into multi-scale attention maps. These maps guide a ConvNeXt-Base encoder, and a lightweight U-Net–style convolutional neural network (CNN) decoder—using nearest-neighbor upsampling and skip connections—reconstructs the final segmentation mask. This design enables the model to focus on semantically relevant regions without relying on full supervision, drastically reducing annotation cost while maintaining high accuracy. We validate our framework on two distinct domains, road crack detection and skin cancer segmentation, demonstrating that it achieves performance comparable to fully supervised segmentation models using only 10–20% of strong annotations. Given the ability of the proposed framework to generalize across varied visual contexts, it has strong potential as a general annotation-efficient segmentation tool for domains where strong labeling is costly or infeasible. Full article

Modern healthcare systems are under growing strain from aging populations, urbanization, and rising chronic disease burdens, creating an urgent need for real-time monitoring and informed decision-making. This survey examines how the convergence of Integrated Sensing and Communication (ISAC) and digital-twin technologies can meet that need by analyzing how ISAC unifies sensing and communication to gather and transmit data with high timeliness and reliability and how digital-twin platforms use these streams to maintain continuously updated virtual replicas of patients, devices, and care environments. Our synthesis compares ISAC frequency options across sub-6 GHz, millimeter-wave, and terahertz bandswith respect to resolution, penetration depth, exposure compliance, maturity, and cost, and it discusses joint waveform design and emerging 6G architectures. It also presents reference architecture patterns that connect heterogeneous clinical sensors to ISAC links, data ingestion, semantic interoperability pipelines using Fast Healthcare Interoperability Resources (FHIR) and IEEE 11073, and digital-twin synchronization, and it catalogs clinical and operational applications, together with validation and integration requirements. We conduct a targeted scoping review of peer-reviewed literature indexed in major scholarly databases between January 2015 and July 2025, with inclusion restricted to English-language, peer-reviewed studies already cited by this survey, and we apply a transparent screening and data extraction procedure to support reproducibility. The survey further reviews clinical opportunities enabled by data-synchronized twins, including personalized therapy planning, proactive early-warning systems, and virtual intervention testing, while outlining the technical, clinical, and organizational hurdles that must be addressed. Finally, we examine workflow adaptation; governance and ethics; provider training; and outcome measurement frameworks such as length of stay, complication rates, and patient satisfaction, and we conclude that by highlighting both the integration challenges and the operational upside, this survey offers a foundation for the development of safe, ethical, and scalable data-driven healthcare models. Full article

The Internet of Things (IoT) has enabled a vast network of devices to communicate over the Internet. However, the fragmentation of IoT systems continues to hinder seamless data sharing and coordinated management across platforms.However, there is currently no actual search engine for IoT data. Existing IoT search engines are considered device discovery tools, providing only metadata about devices rather than enabling access to IoT application data. While efforts such as IoTCrawler have striven to support IoT application data, they have largely failed due to the fragmentation of IoT systems and the heterogeneity of IoT data.To address this, we recently introduced SensorsConnect—a unified framework designed to facilitate interoperable content and sensor data sharing among collaborative IoT systems, inspired by how the World Wide Web (WWW) enabled shared and accessible information spaces for humans. This paper presents the IoT Agentic Search Engine (IoTASE), a real-time semantic search engine tailored specifically for IoT environments. IoTASE leverages LLMs and Retrieval-Augmented Generation (RAG) techniques to address the challenges of navigating and searching vast, heterogeneous streams of real-time IoT data. This approach enables the system to process complex natural language queries and return accurate, contextually relevant results in real time. To evaluate its effectiveness, we implemented a hypothetical deployment in the Toronto region, simulating a realistic urban environment using a dataset composed of 500 services and over 37,000 IoT-like data entries. Our evaluation shows that IoT-ASE achieved 92% accuracy in retrieving intent-aligned services and consistently generated concise, relevant, and preference-aware responses, outperforming generalized outputs produced by systems such as Gemini. These results underscore the potential of IoT-ASE to make real-time IoT data both accessible and actionable, supporting intelligent decision-making across diverse application domains. Full article

Recent studies on detecting hallucinations in summaries follow a method of decomposing summaries into atomic content units (ACUs) and then determining whether each unit logically matches the document text based on natural language inference. However, this fails to consider discourse link relations such as temporal order, causality, and purpose, leading to the inability to detect conflicts in semantic connections between individual summary ACUs, even when the conflicts are present in the document. To overcome this limitation, this study proposes a method of extracting Discourse Link-Aware Content Unit (DL-ACU) by converting the summary into an Abstract Meaning Representation (AMR) graph and structuring the discourse link relations between ACUs. Additionally, to align summary ACUs with corresponding document information in a fine-grained manner, we propose a Selective Document-Atomic Content Unit (SD-ACU). For each summary ACU, the SD-ACU retrieves only the most relevant document sentences and then decomposes them into document ACUs. Applying the DL-ACU module to existing hallucination detection systems such as FIZZ and FENICE reduces the error rate of discourse link errors on FRANK. When both modules are combined, the system improves balanced accuracy and ROC-AUC across major benchmarks. This suggests the proposed method effectively captures discourse link errors while enabling ACU-to-ACU alignment. Full article

With the advancement of image processing techniques, few-shot learning (FSL) has gradually become a key approach to addressing the problem of data scarcity. However, existing FSL methods often rely on unimodal information under limited sample conditions, making it difficult to capture fine-grained differences between categories. To address this issue, we propose a multimodal few-shot learning method based on category name expansion and image feature enhancement. By integrating the expanded category text with image features, the proposed method enriches the semantic representation of categories and enhances the model’s sensitivity to detailed features. To further improve the quality of cross-modal information transfer, we introduce a cross-modal residual connection strategy that aligns features across layers through progressive fusion. This approach enables the fused representations to maximize mutual information while reducing redundancy, effectively alleviating the information bottleneck caused by uneven entropy distribution between modalities and enhancing the model’s generalization ability. Experimental results demonstrate that our method achieves superior performance on both natural image datasets (CIFAR-FS and FC100) and a medical image dataset. Full article

This study investigates the evolution of online public opinion during the COVID-19 pandemic by integrating topic mining with sentiment analysis. To overcome the limitations of traditional short-text models and improve the accuracy of sentiment detection, we propose a novel hybrid framework that combines a GloVe-enhanced Biterm Topic Model (BTM) for semantic-aware topic clustering with a RoBERTa-TextCNN architecture for deep, context-rich sentiment classification. The framework is specifically designed to capture both the global semantic relationships of words and the dynamic contextual nuances of social media discourse. Using a large-scale corpus of more than 550,000 Weibo posts, we conducted comprehensive experiments to evaluate the model’s effectiveness. The proposed approach achieved an accuracy of 92.45%, significantly outperforming baseline transformer-based baseline representative of advanced contextual embedding models across multiple evaluation metrics, including precision, recall, F1-score, and AUC. These results confirm the robustness and stability of the hybrid design and demonstrate its advantages in balancing precision and recall. Beyond methodological validation, the empirical analysis provides important insights into the dynamics of online public discourse. User engagement is found to be highest for the topics directly tied to daily life, with discussions about quarantine conditions alone accounting for 42.6% of total discourse. Moreover, public sentiment proves to be highly volatile and event-driven; for example, the announcement of Wuhan’s reopening produced an 11% surge in positive sentiment, reflecting a collective emotional uplift at a major turning point of the pandemic. Taken together, these findings demonstrate that online discourse evolves in close connection with both societal conditions and government interventions. The proposed topic–sentiment analysis framework not only advances methodological research in text mining and sentiment analysis, but also has the potential to serve as a practical tool for real-time monitoring online opinion. By capturing the fluctuations of public sentiment and identifying emerging themes, this study aims to provide insights that could inform policymaking by suggesting strategies to guide emotional contagion, strengthen crisis communication, and promote constructive public debate during health emergencies. Full article

Intelligent robotic systems frequently operate under stringent energy limitations, especially in complex and dynamic environments. To enhance both adaptability and reliability, this study introduces a semantic planning framework that integrates ontology-driven reasoning with energy awareness. The framework estimates energy consumption based on the platform-specific behavior of sensing, actuation, and computational modules while continuously updating place-level semantic representations using real-time execution data. These representations encode not only spatial and contextual semantics but also energy characteristics acquired from prior operational history. By embedding historical energy usage profiles into hierarchical semantic maps, this framework enables more efficient route planning and context-aware task assignment. A shared semantic layer facilitates coordinated planning for both single-robot and multi-robot systems, with the decisions informed by energy-centric knowledge. This approach remains hardware-independent and can be applied across diverse platforms, such as indoor service robots and ground-based autonomous vehicles. Experimental validation using a differential-drive mobile platform in a structured indoor setting demonstrates improvements in energy efficiency, the robustness of planning, and the quality of the task distribution. This framework effectively connects high-level symbolic reasoning with low-level energy behavior, providing a unified mechanism for energy-informed semantic decision-making. Full article

Weakly supervised semantic segmentation (WSSS) aims to learn pixel-level semantic concepts from image-level class labels. Due to the simplicity and efficiency of the training pipeline, end-to-end WSSS has received significant attention from the research community. However, the coarse nature of pseudo-label regions remains one of the primary bottlenecks limiting the performance of such methods. To address this issue, we propose class-guided enhanced pseudo-labeling (CEP), a method designed to generate high-quality pseudo-labels for end-to-end WSSS frameworks. Our approach leverages pretrained foundation models, such as contrastive language-image pre-training (CLIP) and segment anything model (SAM), to enhance pseudo-label quality. Specifically, following the pseudo-label generation pipeline, we introduce two key components: a Skip-CAM module and a pseudo-label refinement module. The Skip-CAM module enriches feature representations by introducing skip connections from multiple blocks of CLIP, thereby improving the quality of localization maps. The refinement module then utilizes SAM to refine and correct the pseudo-labels based on the initial class-specific regions derived from the localization maps. Experimental results demonstrate that our method surpasses the state-of-the-art end-to-end approaches as well as many multi-stage competitors. Full article

This article examines the “semantic gap” in the digitisation of museum collections—the divide between human-comprehensible representations of artefacts and machine-readable data structures. Drawing on a comparative analysis of national museum databases from Latvia, Estonia, and Finland, the study explores how material objects are transformed into digital surrogates and the challenges of creating interoperable, searchable, and meaningful datasets. Key obstacles include inconsistent metadata standards, linguistic variability, and differences in classification systems, which hinder aggregation and transnational analysis. Case studies of temporal, material, and image-based metadata reveal how human-oriented descriptions—rich in nuance, context, and uncertainty—often resist direct computational translation. The research shows that while digital formats offer powerful opportunities for aggregation, search, and reinterpretation of heritage at scale, this flexibility comes at the cost of reducing object-specific richness. The paper argues that the value, or “aura,” of digitised objects lies in their potential for connectivity and cross-institutional integration, achievable only through metadata standardisation and thoughtful design. Understanding digitisation as a culturally embedded process can help bridge disciplinary perspectives and improve future museum data infrastructures. Full article

As vision and mission statements embody the directions set forth by an organization, their connection to the Sustainable Development Goals (SDGs) must be made explicit to guide overall decision-making in taking strides toward the sustainability agenda. The semantic alignment of these strategic statements with the SDGs is investigated in a previous study, although several limitations need further exploration. Thus, this study aims to advance two contributions: (1) utilizing the capabilities of LLMs (Large Language Models) in text semantic analysis and (2) integrating fuzziness into the problem domain by using a novel intuitionistic fuzzy set extension of the PROBID (Preference Ranking On the Basis of Ideal-average Distance) method. First, a systematic approach evaluates the semantic alignment of organizational strategic statements with the SDGs by leveraging the use of LLMs in semantic similarity and relatedness tasks. Second, viewing it as a multi-criteria decision-making (MCDM) problem and recognizing the limitations of LLMs, the evaluations are represented as intuitionistic fuzzy sets (IFSs), which prompted the development of an IF extension of the PROBID method. The proposed IF-PROBID method was then deployed to evaluate the 47 top Philippine corporations. Utilizing ChatGPT 3.5, 7990 prompts with repetitions generated the membership, non-membership, and hesitance scores for each evaluation. Also, we developed a cohort-dependent SDG–vision–mission matrix that categorizes corporations into four distinct classifications. Findings suggest that “highly-aligned” corporations belong to the private and technology sectors, with some in the industrial and real estate sectors. Meanwhile, “weakly-aligned” corporations come from the manufacturing and private sectors. In addition, case-specific insights are presented in this work. The comparative analysis yields a high agreement between the results and those generated by other IF-MCDM extensions. This paper is the first to demonstrate two methodological advances: (1) the integration of LLMs in MCDM problems and (2) the development of the IF-PROBID method that handles the resulting inherently imprecise evaluations. Full article

Small object detection in UAV imagery remains challenging due to complex aerial perspectives and the presence of dense, small targets with blurred boundaries. To address these challenges, we propose ACD-DETR, an adaptive end-to-end Transformer detector tailored for UAV-based small object detection. The framework introduces three core modules: the Multi-Scale Edge-Enhanced Feature Fusion Module (MSEFM) to preserve fine-grained details; the Omni-Grained Boundary Calibrator (OG-BC) for boundary-aware semantic fusion; and the Dynamic Position Bias Attention-based Intra-scale Feature Interaction (DPB-AIFI) to enhance spatial reasoning. Furthermore, we introduce ACD-DETR-SBA+, a fusion-enhanced variant that removes OG-BC and DPB-AIFI while deploying densely connected Semantic–Boundary Aggregation (SBA) modules to intensify boundary–semantic fusion. This design sacrifices computational efficiency in exchange for higher detection precision, making it suitable for resource-rich deployment scenarios. On the VisDrone2019 dataset, ACD-DETR achieves 50.9% mAP@0.5, outperforming the RT-DETR-R18 baseline by 3.6 percentage points, while reducing parameters by 18.5%. ACD-DETR-SBA+ further improves accuracy to 52.0% mAP@0.5, demonstrating the benefit of SBA-based fusion. Extensive experiments on the VisDrone2019 and DOTA datasets demonstrate that ACD-DETR achieves a state-of-the-art trade-off between accuracy and efficiency, while ACD-DETR-SBA+ achieves further performance improvements at higher computational cost. Ablation studies and visual analyses validate the effectiveness of the proposed modules and design strategies. Full article

High-resolution remote sensing images often suffer from inadequate fusion between global and local features, leading to the loss of long-range dependencies and blurred spatial details, while also exhibiting limited adaptability to multi-scale object segmentation. To overcome these limitations, this study proposes RST-Net, a semantic segmentation network featuring a dual-branch encoder structure. The encoder integrates a ResNeXt-50-based CNN branch for extracting local spatial features and a Shunted Transformer (ST) branch for capturing global contextual information. To further enhance multi-scale representation, the multi-scale feature enhancement module (MSFEM) is embedded in the CNN branch, leveraging atrous and depthwise separable convolutions to dynamically aggregate features. Additionally, the residual dynamic feature fusion (RDFF) module is incorporated into skip connections to improve interactions between encoder and decoder features. Experiments on the Vaihingen and Potsdam datasets show that RST-Net achieves promising performance, with MIoU scores of 77.04% and 79.56%, respectively, validating its effectiveness in semantic segmentation tasks. Full article