Search Results (155)

To address issues commonly observed during the inference phase of large language models—such as inconsistent labels, formatting errors, or semantic deviations—a series of targeted strategies has been proposed. First, a relation label refinement strategy based on semantic similarity and syntactic structure has been designed to calibrate the model’s outputs, thereby improving the accuracy and consistency of label prediction. Second, to meet the contextual modeling needs of different types of instance bags, a multi-level contextual augmentation strategy has been constructed. For multi-sentence instance bags, a graph-based retrieval enhancement mechanism is introduced, which integrates intra-bag entity co-occurrence networks with document-level sentence association graphs to strengthen the model’s understanding of cross-sentence semantic relations. For single-sentence instance bags, a semantic expansion strategy based on term frequency-inverse document frequency is employed to retrieve similar sentences. This enriches the training context under the premise of semantic consistency, alleviating the problem of insufficient contextual information. Notably, the proposed multi-granularity framework captures semantic symmetry between entities and relations across different levels of context, which is crucial for accurate and balanced relation understanding. The proposed methodology offers practical advancements for semantic analysis applications, particularly in knowledge graph development. Full article

Wetlands are critical components of the global carbon cycle, yet their carbon sink dynamics under hydrological fluctuations remain insufficiently understood. This study employed the Carnegie-Ames-Stanford Approach (CASA) model to estimate the net ecosystem productivity (NEP) of the Dongting Lake wetland and explored the spatiotemporal dynamics and driving mechanisms of carbon sinks from 2000 to 2022, utilizing the Theil-Sen median trend, Mann-Kendall test, and attribution based on the differentiating equation (ADE). Results showed that (1) the annual mean spatial NEP was 50.24 g C/m²/a, which first increased and then decreased, with an overall trend of −1.5 g C/m²/a. The carbon sink was strongest in spring, declined in summer, and shifted to a carbon source in autumn and winter. (2) Climate variability and human activities contributed +2.17 and −3.73 g C/m²/a to NEP, respectively. Human activities were the primary driver of carbon sink degradation (74.30%), whereas climate change mainly promoted carbon sequestration (25.70%). However, from 2000–2011 to 2011–2022, climate change shifted from enhancing to limiting carbon sequestration, mainly due to the transition from water storage and lake reclamation to ecological restoration policies and intensified climate anomalies. (3) NEP was negatively correlated with precipitation and water level. Land use adjustments, such as forest expansion and conversion of cropland and reed to sedge, alongside maintaining growing season water levels between 24.06~26.44 m, are recommended to sustain and enhance wetland carbon sinks. Despite inherent uncertainties in model parameterization and the lack of sufficient in situ flux validation, these findings could provide valuable scientific insights for wetland carbon management and policy-making. Full article

Background: Surfactants can be added into polymer–amorphous drug systems to further enhance solubility. However, this may cause amorphous drugs to become physically unstable, and the inherent mechanism at play here is not fully understood. Methods: We explored the effects of poloxamer, a poly (ethylene oxide)-poly (propylene oxide)-poly (ethylene oxide) (PEO-PPO-PEO) triblock copolymer surfactant, and its segments on the nucleation and growth kinetics of amorphous nitrendipine (NTP) from the melt through polarized light microscopy. The effects of poloxamer and structural analogs on the melting point and glass transition temperature were also investigated using differential scanning calorimetry. Results: The poloxamer and its structural analogs enhanced nucleation and growth kinetics in supercooled liquid. Poloxamer and its structural analogs exhibited similar effects on the nucleation and growth kinetics of amorphous NTP, suggesting minimal dependence on structural variation. The overall crystallization rate of the NTP increased when increasing the poloxamer content and ultimately reached a maximum value; after that, the crystallization rates of NTP decreased when increasing the poloxamer content. Conclusions: Poloxamer and its structural analogs achieve similar effects on crystallization due to their comparable plasticizing effects. The nucleation and growth rates show different trends as a function of the poloxamer content. This effect is a result of both kinetic and thermodynamic factors. This study is relevant to understanding the impacts of the surfactant on the physical instability of amorphous drugs. Full article

Background/Objectives: To reduce the early mortality of light-chain amyloidosis (AL), earlier diagnosis is needed. To pursue this goal, we conducted a multicenter study screening for AL λ-type (NCT04615572) in subjects > 60 years of age with λ smoldering myeloma (SMM) or monoclonal gammopathy of undetermined significance (MGUS), a light-chain differential (dFLC, λ minus κ) > 23 mg/L, and no prior amyloid diagnosis. Methods: Variables included AL-related IGVL gene usage and clonal plasma cell cytogenetic abnormalities, such as t(11;14) or gain 1q, which are present in 75% of AL cases. Here, 9 out of 33 λ IGVL genes, accounting for 90% of AL λ cases, were considered to be AL-related. Bone marrow was obtained, plasma cell cytogenetics and next generation sequencing for IGVL genes were performed, and subjects with AL-related IGVL genes were screened for AL using tissue studies. Results: From 2021 to 2023, we enrolled 30 subjects (19 M/11 F) with a median age of 68.5 years old (IQR 64.3–73), 17 SMM and 13 MGUS, with a median of 6% marrow plasma cells (range, 3.5–40). Here, 11 SMM and 4 MGUS cases had t(11;14) or gain 1q; 10/17 SMM and 12/13 MGUS had AL-related genes, and AL was ultimately confirmed by tissue biopsy in 3 with SMM. SMM, AL-related IGVL genes, and t(11;14) or gain 1q were found in 6 SMM subjects, including the 3 with AL (3/6 vs. 0/16; p < 0.05, Fisher’s exact, two-tailed). Conclusions: These results justify a larger study screening for AL in SMM to develop a likelihood algorithm for AL using dFLC, IGVL gene usage, and the presence of t(11;14) or gain 1q. Full article

The sloshing in storage tanks can exert negative influences on the safety and stability of tank structures undergoing earthquake excitation. An analytical mechanical model is presented here to investigate the seismic responses of a base-isolated cylindrical tank resting on soil. The continuous liquid sloshing is modeled as the convective spring–mass, the impulsive spring–mass, and the rigid mass. The soil impedances are equivalent to the systematic lumped-parameter models. The bearing isolation is considered as the elastic–viscous damping model. A comparison between the present and reported results is presented to prove the accuracy of the coupling model. A parametric analysis is carried out for base-isolated broad and slender tanks to examine the effects of the isolation period, isolation damping ratio, tank aspect ratio, and soil stiffness on structural responses. The results show that the interaction between soft soil and the base-isolated tank exerts significant influence on earthquake responses. Full article

Hydrogeochemical characteristics and temperature variations in fault-controlled, deep-circulation thermal springs elucidate water–rock interaction dynamics and hydrothermal circulation depths, providing critical insights into fault permeability and stress accumulation. To investigate the coexistence of high-temperature and medium-low-temperature thermal springs on small-scale faults and their distinct circulation mechanisms, hydrochemical and isotopic analyses were conducted on 13 water samples (9 proximal on the Xiangbaihe Fault) in western Yunnan. The hot springs along the Xiangbaihe Fault are predominantly classified as the Na-HCO₃ type, derived from carbonate and aluminosilicate hydrolysis. δ²H and δ¹⁸O confirmed a meteoric origin, with recharge elevations spanning 2465–3286 m (Gaoligong Mountain). Inverse hydrochemical modeling demonstrated progressive mineral transfer and water–rock interactions along the fault’s east–west axis. Conservative elements (Cl, Li) suggested a shared geothermal fluid source or reservoir affiliation. BLZ reservoir temperatures (194–221 °C) were classified as a high-temperature system, whereas others (58–150 °C) represented medium-low-temperature systems. Although each thermal spring represents a distinct geothermal system, reservoir interconnectivity is inferred. Notably, despite uniform lithology, variations in spring temperature and elemental composition are attributed to a subsurface magma chamber beneath BLZ, heterogeneous fault geometries, differential reservoir temperatures, and variable cold-water mixing ratios. This study establishes a framework for understanding groundwater circulation in small-scale fault-associated geothermal systems, with implications for tectonic activity monitoring and geothermal resource assessment. Full article

Crack defects pose a significant threat to the safe operation of the foundational structures of hydropower stations. Therefore, crack detection in a plunge pool corridor is crucial for the safe operation and maintenance of hydropower hubs. Addressing challenges such as missing crack datasets, complex and diverse crack shapes, and difficulties in feature extraction, this paper proposes an improved U-Net model. The model integrates the CDIAM attention module within the U-Net architecture and is optimized through the incorporation of the MIDConv convolutional block. The enhanced U-Net model demonstrates outstanding performance on a self-constructed plunge pool corridor crack dataset, achieving an MIoU, F1-score, MPA, and accuracy of 84.44%, 93.05%, 89.75%, and 99.46%, respectively. Compared to the original U-Net model, these metrics show improvements of 5.34%, 4.90%, 3.39%, and 0.36%, respectively. The improved U-Net model exhibits higher detection accuracy in corridor crack segmentation tasks, effectively extracting feature information from irregular shapes, and provides a superior solution for crack defect detection in a plunge pool corridor, thereby enhancing the safety and efficient maintenance of hydropower stations. Full article

With the increasing severity of climate change, Carbon Capture, Utilization, and Storage (CCUS) technology has become essential for reducing atmospheric CO₂. Marine carbon sequestration, which stores CO₂ in seabed geological structures, offers advantages such as large storage capacity and high stability. Cryogenic hoses are critical for the ship-to-ship transfer of liquid CO₂ from transportation vessels to offshore carbon sequestration platforms, but their design methods and mechanical analysis remain inadequately understood. This study reviews existing cryogenic hose designs, including reinforced corrugated hoses, vacuum-insulated hoses, and composite hoses, to assess their suitability for liquid CO₂ transfer. Based on CO₂’s physicochemical properties, a conceptual composite hose structure is proposed, featuring a double-spring-supported internal composite hose, thermal insulation layer, and outer sheath. Practical recommendations for material selection, corrosion prevention, and monitoring strategies are provided to improve flexibility, pressure resistance, and thermal insulation, enabling reliable long-distance tandem transfer. A mechanical analysis framework is developed to evaluate structural performance under conditions including mechanical loads, thermal stress, and dynamic responses. This manuscript includes an introduction to the background, the methodology for data collection, a review of existing designs, an analysis of CO₂ characteristics, the proposed design methods, the mechanical analysis framework, a discussion of challenges, and the conclusions. Full article

Background: Liver regeneration is a critical measure of liver health and plays an essential role in inhibiting the progression of fibrotic lesions and preventing liver failure after hepatocellular carcinoma surgery. However, there are no approved drugs to address this clinical challenge. Methods: The effects of TUDCA on liver regeneration and fibrosis were studied using BRL-3A cells, a partial hepatectomy (PH) rat liver regeneration model, and a carbon tetrachloride (CCl₄)-induced liver fibrosis model. GATA3-knockdown BRL-3A cells were employed to assess the role of GATA3 in TUDCA-induced proliferation. Results: TUDCA promoted the proliferation of BRL-3A cells and enhanced liver regeneration in PH rats while ameliorating liver fibrosis in CCl₄-treated rats. Additionally, the knockdown of GATA3 eliminated the proliferative effect of TUDCA on BRL-3A cells. Conclusions: TUDCA promotes liver regeneration and alleviates liver fibrosis by activating GATA3. Full article

In this study, a total of 26 groundwater samples were collected from the northwest of the Sichuan Basin. Statistical analysis revealed that Ca²⁺ was the predominant cation, followed by Na⁺, Mg²⁺, and K⁺. The anion concentrations followed the order HCO₃⁻ > SO₄²⁻ > NO₃⁻ > F⁻ > Cl⁻. Consequently, Ca-HCO₃ was identified as the dominant hydrochemical type in the study area. Geochemical modeling results indicated that silicate weathering and cation exchange processes were the primary factors influencing groundwater hydrochemistry. To provide an accurate assessment of water quality, a Comprehensive Water Quality Index (CWQI) was applied in this study. This novel method combined factor analysis and the entropy-weighted approach to derive integrated weights for water quality calculation. The CWQI results showed that 73.08% of the samples were classified as excellent for drinking, while 26.92% were classified as good. Sensitivity analysis further demonstrated the robustness of the drinking water quality model. The findings of this study could contribute to the enhancement of water quality evaluation in the Sichuan Basin. Full article

Extreme rainfall events are significant manifestations of climate change, causing substantial impacts on urban infrastructure and public life. This study takes the extreme rainfall event in Beijing in 2023 as the background and utilizes data from Sina Weibo. Based on large language models and prompt engineering, disaster information is extracted, and a multi-factor coupled disaster multi-sentiment classification model, Bert-BiLSTM, is designed. A disaster analysis framework focusing on three dimensions of theme, location and sentiment is constructed. The results indicate that during the pre-disaster stage, themes are concentrated on warnings and prevention, shifting to specific events and rescue actions during the disaster, and post-disaster, they express gratitude to rescue personnel and highlight social cohesion. In terms of spatial location, the disaster shows significant clustering, predominantly occurring in Mentougou and Fangshan. There is a clear difference in emotional expression between official media and the public; official media primarily focuses on neutral reporting and fact dissemination, while public sentiment is even richer. At the same time, there are also variations in sentiment expressions across different affected regions. This study provides new perspectives and methods for analyzing extreme rainfall events on social media by revealing the evolution of disaster themes, the spatial distribution of disasters, and the temporal and spatial changes in sentiment. These insights can support risk assessment, resource allocation, and public opinion guidance in disaster emergency management, thereby enhancing the precision and effectiveness of disaster response strategies. Full article

Essential oils serve as potential additives that can enhance immune respons and disease resistance and regulate the gut microbiota of fish. Here, this research aims to identify the impacts of eucalyptus essential oil (EEO) on growth, liver antioxidative and immune parameters, resistance to Streptococcus iniae, intestinal morphology and gut microbiota in Trachinotus ovatus. All fish (initial weight: 26.87 ± 0.30 g) were randomly allocated to 12 floating cages (2.0 × 2.0 × 2.0 m³) with each cage containing 100 fish and fed for 30 days. Four diets were manufactured with the supplementation of varying levels of EEO (control and 5.0, 10.0, and 15.0 mL/kg) and were named CG, EEO1, EEO2 and EEO3, respectively. The results showed that EEO1 and EEO2 diets significantly increased WGR, thickness of the intestinal muscle layer, and the ratio of villus height to crypt depth (V/C), while decreasing the intestinal crypt depth of T. ovatus (p < 0.05). Significantly increased activities of SOD and CAT and significantly reduced MDA levels were present in the EEO1 and/or EEO2 groups (p < 0.05). Moreover, the mRNA levels of nrf2, HO-1, GSH-Px, SOD, C4 and GR genes were significantly upregulated and the expression of keap1 and HSP70 genes were significantly downregulated within the EEO1 and EEO2 groups (p < 0.05). After challenge with S. iniae B240703 for 24 h, the bacterial load for five organs in the EEO2 group was less than that in the CG group (p < 0.05). In addition, the fish fed EEO1 and/or EEO2 diets had significantly lower abundances of pathogenic bacteria (Proteobacteria, Planctomycetota, Burkholderia-Caballeronia-Paraburkholderia, Pseudomonas and Blastopirellula) and a higher beneficial bacteria proportion (Firmicutes) than those fed the CG diets (p < 0.05). In conclusion, a moderate dietary m EO level (5.0~10.0 mL/kg) improved the growth and gut morphology, promoted liver immune response, enhanced resistance to S. iniae and modulated the gut microbiota of T. ovatus. Full article

In this study, an acid–base proton transfer method was used to fabricate functional coatings on Mg surfaces with the cooperative effect of durable superhydrophobic and exceptional self-cleaning properties, providing high-efficiency corrosion protection. In this context, aluminum chloride served as a precursor for the direct growth of aluminum hydroxide on the Mg surface. Without the addition of any solvent, the densely arranged absolute palmitic acid was strongly bonded to the grown aluminum hydroxide on the Mg substrate, which acted as an effective anti-water barrier that can impede the penetration of water, as well as the oxygen and chloridion involved. Full article

Data with time attributions such as price, load, and stock, which directly reflect the variation tendency, are the most common type of data character available. However, it is difficult to predict complex and volatile time-series character data. Further, most density cluster methods employ existing data to train the initial radius; however, a certain density radius is hard to be made suitable for continuously generated on-going datasets. Therefore, how to select a suitable timespan according to the time-series character in a way that makes it possible to support an adaptive updated density radius for real-time calculation is a core process. In this paper, a self-adaptive multi-density (SAMD) prediction model is proposed for solving the dynamic density radius selection problem in time-series data so as to improve the accuracy of real-time prediction. This multi-density clustering method can effectively shorten the iteration times and achieve dynamic clustering by the proposed jump sequence, which can optimize the jump points in the electricity price sequence. Moreover, we especially focus on the time interval features and other multi-source influencing factors together to construct the multi-core function with double-layer optimization to calculate the weighted coefficients, which have good adaptability and improve the classification and recognition performance. The experimental results show that the model had higher prediction accuracy and reduced processing time consumption in order to achieve real-time prediction. Full article

The accurate extraction of individual tree positions is key to forest structure quantification, and Unmanned Aerial Vehicle (UAV) visible light data have become the primary data source for extracting individual tree locations. Compared to deep learning methods, classical detection methods require lower computational resources and have stronger interpretability and applicability. However, in closed-canopy forests, challenges such as crown overlap and uneven light distribution hinder extraction accuracy. To address this, the study improves the existing Revised Local Maxima (RLM) method and proposes a Multi-Source Local Maxima (MSLM) method, based on UAV visible light data, which integrates Canopy Height Models (CHMs) and Digital Orthophoto Mosaics (DOMs). Both the MSLM and RLM methods were used to extract individual tree positions from three different types of closed-canopy stands, and the extraction results of the two methods were compared. The results show that the MSLM method outperforms the RLM in terms of Accuracy Rate (85.59%), Overall Accuracy (99.09%), and F1 score (85.21%), with stable performance across different forest stand types. This demonstrates that the MSLM method can effectively overcome the challenges posed by closed-canopy stands, significantly improving extraction precision. These findings provide a cost-effective and efficient approach for forest resource monitoring and offer valuable insights for forest structure optimization and management. Full article