Search Results (1,006)

Recently, music generation models based on deep learning have made remarkable progress in the field of symbolic music generation. However, the existing methods often have problems of violating musical rules, especially since the control of harmonic structure is relatively weak. To address these limitations, this paper proposes a novel framework, the Entropy-Regularized Latent Diffusion for Harmony-Constrained (ERLD-HC), which combines a variational autoencoder (VAE) and latent diffusion models with an entropy-regularized conditional random field (CRF). Our model first encodes symbolic music into latent representations through VAE, and then introduces the entropy-based CRF module into the cross-attention layer of UNet during the diffusion process, achieving harmonic conditioning. The proposed model balances two key limitations in symbolic music generation: the lack of theoretical correctness of pure algorithm-driven methods and the lack of flexibility of rule-based methods. In particular, the CRF module learns classic harmony rules through learnable feature functions, significantly improving the harmony quality of the generated Musical Instrument Digital Interface (MIDI). Experiments on the Lakh MIDI dataset show that compared with the baseline VAE+Diffusion, the violation rates of harmony rules of the ERLD-HC model under self-generated and controlled inputs have decreased by 2.35% and 1.4% respectively. Meanwhile, the MIDI generated by the model maintains a high degree of melodic naturalness. Importantly, the harmonic guidance in ERLD-HC is derived from an internal CRF inference module, which enforces consistency with music-theoretic priors. While this does not yet provide direct external chord conditioning, it introduces a form of learned harmonic controllability that balances flexibility and theoretical rigor. Full article

Efficient emergency load transfer is crucial for ensuring the power system’s safe operation and reliable power supply. However, traditional load transfer methods that rely on human experience have limitations, such as slow response times and low efficiency, which make it difficult to address complex and diverse fault scenarios effectively. Therefore, this paper proposes an emergency load transfer method based on knowledge graphs to achieve intelligent management and efficient retrieval of emergency knowledge. Firstly, a named entity recognition model based on ERNIE-BiGRU-CRF is constructed to automatically extract key entities and relationships from the load transfer plan texts, obtaining information such as fault names, fault causes, and operation steps. Secondly, a power system emergency load transfer knowledge graph is constructed based on the extracted structured knowledge, which is efficiently stored using a graph database and enables the visualization and interactive query of knowledge. Finally, real power system fault cases prove that the proposed method can effectively improve the retrieval efficiency of fault knowledge and provide intelligent support for online emergency load transfer decisions. Full article

The development of controlled-release fertilizers (CRFs) has faced significant challenges due to high hydrophilicity and short release lifespan of bio-based materials, as well as non-renewable and high cost of polyester polyols (PPs). In this study, lignin-based polyols (LPs) and PPs were modified to form a cross-linked polymer film on the surface of urea through an in situ reaction. This approach effectively balanced the slow-release ability and environmental protection of controlled-release fertilizer films. A two-factor, five-level orthogonal test was designed for the mass ratio of lignin/polyester polyol and polyol/polyaryl polymethylene isocyanate (PAPI), comprising a total of 25 treatments. The results indicated that the appropriateness of lignin polyols increased the hydrogen bond content of polyurethane membrane, improved the mechanical strength of the fertilizer membrane shell, and effectively reduced friction losses during storage and transportation. Moreover, optimizing the polyol-to-PAPI ratio minimized coating porosity, produced a smoother and denser surface, and prolonged the nitrogen release period. When the lignin polyol dosage was 25% and the polyol to PAPI ratio was 1:2, the nitrogen release time of the prepared coated urea extended to 32 days, which was 3.5 times longer than that of lignin polyurethane coated urea (7 days). The incorporation of lignin and the optimal ratio of coating materials significantly improved the controlled-release efficiency of coated fertilizer, providing theoretical support for the sustainable agricultural application of biomass. Full article

Studies of HIV-1 molecular epidemiology describe significant differences in HIV infection spread across geographical areas. We examined 80 HIV-1 samples from the Jewish Autonomous Region of Russia in 2024. HIV-1 genome sequences for 12 samples revealed a novel HIV-1 called CRF173_63A6. HIV-1 CRF173_63A6 was found to have arisen through recombination between a specific Russian A6 subtype and the recombinant virus CRF63_02A6, which is responsible for the PWID-associated HIV outbreak in the Siberian region of Russia. Phylogenetic analysis of pol sequences previously deposited in Genbank showed that the CRF173_63A6 samples we described are grouped into a common phylogenetic cluster that includes 54 HIV-1 samples isolated in the JAR and other areas of the Russian Far East, indicating a wide distribution of this virus genovariant. This study once again proves the significant contribution of the key PWID group not only to the development of local Russian HIV epidemics, but also to the change in the characteristics of the circulating virus population. Full article

Hyperspectral classification of wetland environments remains challenging due to high spectral similarity, class imbalance, and blurred boundaries. To address these issues, we propose a novel Multi-Branch Channel-Gated Swin Transformer network (MBCG-SwinNet). In contrast to previous CNN-based designs, our model introduces a Swin Transformer spectral branch to enhance global contextual modeling, enabling improved spectral discrimination. To effectively fuse spatial and spectral features, we design a residual feature interaction chain comprising a Residual Spatial Fusion (RSF) module, a channel-wise gating mechanism, and a multi-scale feature fusion (MFF) module, which together enhance spatial adaptivity and feature integration. Additionally, a DenseCRF-based post-processing step is employed to refine classification boundaries and suppress salt-and-pepper noise. Experimental results on three UAV-based hyperspectral wetland datasets from the Yellow River Delta (Shandong, China)—NC12, NC13, and NC16—demonstrate that MBCG-SwinNet achieves superior classification performance, with overall accuracies of 97.62%, 82.37%, and 97.32%, respectively—surpassing state-of-the-art methods. The proposed architecture offers a robust and scalable solution for hyperspectral image classification in complex ecological settings. Full article

The growing development of conductive functionalised textiles has attracted the interest of the automotive industry, which is seeking innovative solutions for seamless and futuristic interior design aimed at improving both vehicle aesthetics and user experience. In line with this trend, the present work investigates the electrical performances of two conductive flocked yarns, one incorporating silver-coated fibres and the other carbon black-based fibres, for potential application in smart automotive interiors. The stability of their electrical properties was also evaluated under thermal ageing and mechanical stress conditions. Thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), and field emission scanning electron microscopy (FE-SEM) investigations provided information about the composition and structural properties of the yarns. Silver-based yarns demonstrated superior conductivity and thermal stability. In contrast, carbon-black yarns exhibited lower electrical performance and increased sensitivity to ageing due to filler agglomeration. A multitouch capacitive sensor prototype was also developed using the silver-based fabric and successfully integrated into a microcontroller platform. The results demonstrate the suitability of conductive flocked textiles for durable, low-voltage human–machine interfaces requiring robust, flexible, and responsive textile-based control surfaces, such as automotive applications, consumer electronics, and wearable technology. Full article

Nowadays, with the accumulation of large amounts of stress-response transcriptomic data in plants, it is possible to clarify the key genes and transcription factors (TFs) involved in these processes. Here, we present the comprehensive transcriptomic meta-analysis of the high light (HL) response in photosynthetic tissues of Arabidopsis thaliana (L.) Heynh., offering new insights into adaptation mechanisms of plants to excessive light and involved gene regulatory networks. We analyzed 21 experiments covering 58 HL conditions in total, yielding 218,000 instances of differentially expressed genes (DEGs) corresponding to 19,000 unique genes. Based on these data, we developed the publicly accessible AraLightDEGs resource, which offers multiple search filters for experimental conditions and gene characteristics, and we conducted a detailed meta-analysis using our R pipeline, AraLightMeta. Our meta-analysis highlighted distinct transcriptional programs between short- and long-term HL responses in leaves, revealing novel regulatory interactions and refining the understanding of key DEGs. In particular, long-term HL adaptation involves key TFs such as CRF3 and PTF1 regulating antioxidant and jasmonate signaling; ATWHY2, WHY3, and emb2746 coordinating chloroplast and mitochondrial gene expression; AT2G28450 governing ribosome biogenesis; and AT4G12750 controlling methyltransferase activity. We integrated these findings into a conceptual scheme illustrating transcriptional regulation and signaling processes in leaf cells responding to long-term HL stress. Full article

Loess landslide disasters are influenced by a multitude of factors, including slope conditions, triggering mechanisms, and spatial attributes. Extracting these factors from unstructured geological texts is challenging due to nested entities, semantic ambiguity, and rare domain-specific terms. This study proposes a joint extraction framework guided by a domain ontology that categorizes six types of loess landslide influencing factors, including spatial relationships. The ontology facilitates conceptual classification and semi-automatic nested entity annotation, enabling the construction of a high-quality corpus with eight tag types. The model integrates a Soft-Lexicon mechanism that enhances character-level GloVe embeddings with explicit lexical features, including domain terms, part-of-speech tags, and word boundary indicators derived from a domain-specific lexicon. The resulting hybrid character-level representations are then fed into a BiLSTM-CRF architecture to jointly extract entities, attributes, and multi-level spatial and causal relationships. Extracted results are structured using a content-knowledge model to build a spatially enriched knowledge graph, supporting semantic queries and intelligent reasoning. Experimental results demonstrate improved performance over baseline methods, showcasing the framework’s effectiveness in geohazard information extraction and disaster risk analysis. Full article

Volunteer firefighters often have lower cardiorespiratory fitness (CRF) and less access to health monitoring and fitness programs than career firefighters, yet few studies explore how individual and departmental factors influence their CRF. This study assessed associations between CRF and both firefighter-level (e.g., years of service, firefighting calls, and firefighter rank) and department-level (e.g., department characteristics and fitness infrastructure) factors among volunteer firefighters. Surveys were administered to United States volunteer firefighters and departments, capturing CRF and related characteristics. CRF was analyzed as both a continuous and categorical variable (≤8, >8–<10, 10–<12, ≥12 METs) using bivariate analyses and mixed effects linear and logistic regression. Among 569 incumbent volunteer firefighters from 41 departments, 79.9% did not meet the recommended 12 METs threshold. Only 56.8% of departments provided routine physical exams; 35.1% had a wellness coordinator or committee; and 40.5% offered fitness resources. More years of service were associated with lower CRF and reduced odds of meeting the 12 METs benchmark, while more frequent training and responding to more calls were associated with better CRF. These findings highlight individual and structural challenges for CRF in volunteer fire service, underscoring the need for targeted fitness support to protect firefighter health and community safety. Full article

Conventional unconditional random field (URF) models were shown to neglect in-situ monitoring data and thus misrepresent real slope stability. To address this, a conditional random field (CRF) generator was proposed, in which Fast Fourier Transform (FFT) filtering was coupled with co-Kriging to assimilate site observations. A representative three-bench slope was adopted, and the failure-mode distribution and the statistics of the factor of safety (FoS) produced by the URF, the independent random field (IRF), and the CRF were examined across bedding-dip angles of 15–75° and two cross-correlation states ($\rho$ = −0.2, 0). It was found that eliminating cross-correlation decreased the mean FoS by 0.006, increased its standard deviation by 10.26%, and raised the frequency of low-FoS events from 7.49% to 12.30%. When field constraints were imposed through the CRF, the probability of through-going failure was reduced by 12%, the mean FoS was increased by 0.01, the standard deviation was reduced by 15.38%, and low-FoS events were suppressed to 2.30%. The CRF framework was thus demonstrated to integrate stochastic analysis with field measurements, enabling more realistic reliability assessment and proactive risk management of slopes. Full article

Gas leak detection in oil and gas processing facilities is a critical component of the safety production monitoring system. Non-contact detection technology based on infrared imaging has emerged as a vital real-time monitoring method due to its rapid response and extensive coverage. However, existing pixel-level segmentation networks face challenges such as insufficient segmentation accuracy, rough gas edges, and jagged boundaries. To address these issues, this study proposes a novel pixel-level segmentation network training framework based on anchor box annotation and enhances the segmentation performance of the YOLOv8-seg network for gas detection applications. First, a dynamic threshold is introduced using the Visual Background Extractor (ViBe) method, which, in combination with the YOLOv8-det network, generates binary masks to serve as training masks. Next, a segmentation head architecture is designed, incorporating dynamic kernels and multi-branch collaboration. This architecture utilizes feature concatenation under deformable convolution and attention mechanisms to replace parts of the original segmentation head, thereby enhancing the extraction of detailed gas features and reducing dependency on anchor boxes during segmentation. Finally, a joint Dice-BCE (Binary Cross-Entropy) loss, weighted by ViBe-CRF (Conditional Random Fields) confidence, is employed to replace the original Seg_loss. This effectively reduces roughness and jaggedness at gas edges, significantly improving segmentation accuracy. Experimental results indicate that the improved network achieves a 6.4% increase in F1 score and a 7.6% improvement in the mIoU (mean Intersection over Union) metric. This advancement provides a new, real-time, and efficient detection algorithm for infrared imaging of gas leaks in oil and gas processing facilities. Furthermore, it introduces a low-cost weakly supervised learning approach for training pixel-level segmentation networks. Full article

Ratoon rice is a planting system that efficiently utilizes temperature and light resources. However, multiple fertilization applications are typically required to maintain stable rice yields. Improper fertilization not only poses challenges to scarce labor resources but also increases carbon footprints (CFs). Research on the effects of different fertilization strategies on greenhouse gas (GHG) emissions, yield, CF, and ecosystem net economic benefits (NEEBs) in ratoon rice systems remains limited. A two-year field experiment was conducted to evaluate the effects of one conventional fertilization strategy and four optimized fertilization strategies on GHG emissions, yield, CF, and NEEBs in the ratoon rice system. The conventional fertilization strategy applied urea in five splits (FFP, 280 kg N·ha⁻¹). The optimized strategies included (1) one-time side deep application controlled-release fertilizer (CRF, 280 kg N·ha⁻¹); (2) CRF with 20% N replaced by organic fertilizer (OF + CRF1); (3) the same as (2) with a 10% N reduction (OF + CRF2, 252 kg N·ha⁻¹); and (4) the same as (2) with a 20% N reduction (OF + CRF3, 224 kg N·ha⁻¹). The results showed that compared with FFP, optimized fertilization treatments reduced CH₄ and N₂O emissions by 28.69% to 55.27% and 25.08% to 40.32%, respectively. They also increased the annual rice yields by 2.22% to 19.52% (except OF + CRF3). Optimizing fertilization treatments reduced annual CF, CF_Y, and CF_EC by 26.66% to 49.59%, 34.11% to 51.12%, and 25.35% to 41.47%, respectively. These treatments also increased NEEBs by 8.27% to 34.23%. Among them, OF + CRF1 and OF + CRF2 treatments achieved the highest NEEB. In summary, CRF treatments can balance ratoon rice yield and environmental benefits. Replacing part of the N with organic fertilizer further enhances annual yield and NEEBs. Full article

Dietary reviews contain rich emotional and objective information; however, existing knowledge extraction methods struggle with low-resource scenarios due to sparse and imbalanced label distributions. To address these challenges, this paper proposes a few-shot supervised learning approach. First, we develop a professional dietary–emotional schema by integrating domain knowledge with real-time data to ensure the coverage of diverse emotional expressions. Next, we introduce a dataset optimization method based on dual constraints—label frequency and quantity—to mitigate label imbalance and improve model performance. Utilizing the optimized dataset and a tailored prompt template, we fine-tune the DRE-UIE model for multivariate knowledge extraction. The experimental results demonstrate that the DRE-UIE model achieves a 20% higher F1 score than BERT-BiLSTM-CRF and outperforms TENER by 1.1%. Notably, on a 20-shot subset, the model on the Chinese dataset scores 0.841 and attains a 15.16% F1 score improvement over unoptimized data, validating the effectiveness of our few-shot learning framework. Furthermore, the approach also exhibits robust performance across Chinese and English corpora, underscoring its generalization capability. This work offers a practical solution for low-resource dietary–emotional knowledge extraction by leveraging schema design, dataset optimization, and model fine-tuning to achieve high accuracy with minimal annotated data. Full article

Knowledge graph technology holds significant importance for efficient fault diagnosis in household appliances. However, the scarcity of public fault diagnosis data and the lack of automated knowledge extraction pose major challenges to knowledge graph construction. To address issues such as ambiguous entity boundaries, severe entity nesting, and poor entity extraction performance in fault diagnosis texts, this paper proposes a dual-strategy progressive knowledge extraction framework. First, to tackle the high complexity of fault diagnosis texts, an entity recognition model named RoBERTa-zh-BiLSTM-MUL-CRF is designed, improving the accuracy of nested entity extraction. Second, leveraging the semantic understanding capability of large language models, a progressive prompting strategy is adopted for ontology alignment and relation extraction, achieving automated knowledge extraction. Experimental results show that the proposed named entity recognition model outperforms traditional models, with improvements of 3.87%, 5.82%, and 2.05% in F1-score, recall, and precision, respectively. Additionally, the large language model demonstrates better performance in ontology alignment compared to traditional machine learning models. The constructed knowledge graph for household appliance fault diagnosis integrates structured fault diagnosis information. It effectively processes unstructured fault texts and supports visual queries and entity tracing. This framework can assist maintenance personnel in making rapid judgments, thereby improving fault diagnosis efficiency. Full article

As cyberattacks continue to rise in frequency and sophistication, extracting actionable Cyber Threat Intelligence (CTI) from diverse online sources has become critical for proactive threat detection and defense. However, accurately identifying complex entities from lengthy and heterogeneous threat reports remains challenging due to long-range dependencies and domain-specific terminology. To address this, we propose XLNet-CRF, a hybrid framework that combines permutation-based language modeling with structured prediction using Conditional Random Fields (CRF) to enhance Named Entity Recognition (NER) in cybersecurity contexts. XLNet-CRF directly addresses key challenges in CTI-NER by modeling bidirectional dependencies and capturing non-contiguous semantic patterns more effectively than traditional approaches. Comprehensive evaluations on two benchmark cybersecurity corpora validate the efficacy of our approach. On the CTI-Reports dataset, XLNet-CRF achieves a precision of 97.41% and an F1-score of 97.43%; on MalwareTextDB, it attains a precision of 85.33% and an F1-score of 88.65%—significantly surpassing strong BERT-based baselines in both accuracy and robustness. Full article