As a basic task in natural language processing, named entity recognition (NER) can effectively identify and classify named entities in text. Some progress has been made in entity recognition with span-based methods, but the quality differences between candidate spans are often overlooked. To tackle the problem, a named entity recognition method that fuses boundary interaction information is proposed. A boundary interaction module is used to evaluate the semantic associations and interaction strengths between boundaries, and a boundary interaction information matrix is generated. This matrix is used to identify potential semantic connections between boundaries, guiding the model to recognize and mark high-quality candidate spans. Additionally, a multi-scale dilated convolution module is integrated to reduce the impact of non-entity noise by utilizing the semantic relationships between spans. It is demonstrated through experiments that the method achieves F₁ scores of 89.78%, 87.37%, and 72.10% on the ACE2005 Chinese dataset, ACE2005 English dataset, and Weibo dataset, respectively. These results represent improvements of 0.67, 0.95, and 0.69 percentage points over baseline models, validating the effectiveness of the proposed method for named entity recognition.

Relation extraction aims to extract the semantic relationship between two named entities. Recently, prompt learning has unified the optimization objectives of pre-trained language models and fine-tuning by concatenating prompt templates and performing mask prediction, achieving excellent performance in the field of relationship extraction. However, weak semantic associations between fixed prompt templates and relation instances are observed, which limits the model’s ability to perceive complex relationships. To address this issue, a relation extraction method based on mask attention and multi-feature convolution networks is proposed. The tri-affine attention mechanism is adopted to interactively map the mask in the prompt template with the semantic space of the original text. Two-dimensional mask semantics are then formed through this process. Multi-feature convolutional networks and multi-layer perceptron are employed to extract relational information from the two-dimensional mask semantics. Explicit semantic dependencies between the mask and the relation instance are established. This approach enhances the semantic perception of complex relationships in prompt models. Performances of 91.4%, 91.2%, and 82.6% are achieved on the SemEval, SciERC, and CLTC datasets, respectively. The effectiveness of the proposed method is demonstrated by experimental results.

Zero-shot entity retrieval, which aims to link mentions to entities unseen during training, plays a vital role in many natural language processing tasks. However, previous methods suffer from two main limitations: (1) The use of only the first k sentences of entity descriptions to construct multi-view representations leads to redundancy and loss of semantic information in these views, making it difficult to fully learn the matching relationship between mentions and entities; (2) The focus solely on mentions to construct positive and negative examples, with inadequate consideration of the comparative relationships between mentions and entities, results in incorrect matchings. To address these issues, a topic-based multi-view entity representations (Topic-MVER) method is proposed in this paper. This method constructs multi-view representations for entities based on topics and employs contrastive learning to model three types of relationships between mentions and entities, enhancing the matching degree between them. Finally, the method achieves Recall@1 scores of 48.13% and 73.86% on the ZESHEL and MedMentions datasets, respectively, presenting improvements of 2.73% and 1.21% over the baseline models. This validates the effectiveness of the proposed method.

The influence of fiber coating on the temperature sensitivity of Long period fiber grating (LPFG) is theoretically analyzed. The transmission spectra of LPFG with and without a coating layer are simulated using numerical analysis, and the mode coupling process of LPFG for temperature sensing is analyzed. The results show that the larger the cladding mode order, the greater the effect of the coating layer on the effective refractive index of different cladding modes. Then it is inferred that the LPFG resonance peaks associated with different cladding modes have different temperature sensitivities. The coating layer protects the optical fiber and improves its mechanical strength. Still, more importantly, the coating layer can effectively enhance the temperature sensitivity of the transmission peaks coupled with higher-order cladding modes. At the same time, the results are useful for experimental studies of gratings prepared by retaining the coating layer and polymer-coated grating structures.

To improve the classification performance of motor imagery (MI) tasks by deeply mining and using the characteristic information of electroencephalogram (EEG) signals has always been the focus of brain-computer interfaces (BCI) research. Because EEG feature space is highly dimensional and directly related to both amplitude and phase of EEG signals, how to simultaneously represent and utilize the information contained in amplitude and phase has become a difficult issue. To address this issue, a three-dimensional complex convolutional neural network based on complex-value covariance features is proposed. Firstly, complex-value covariance matrices related to different frequencies as MI-based EEG features is constructed. As a result, complex value can combine the amplitude and phase information of EEG signals together. Moreover, the covariance matrices can preserve multivariate information such as amplitude, phase, spatial locations, frequency, etc. required for classification. Secondly, a full complex convolutional neural network is designed for learning the covariance features and thus achieving high performance classification. Experimental results on two publicly available datasets show that the proposed method can achieve mean accuracies that are at least 2.49 and 1.85 percentage points higher than state-of-the-art methods.

Feature decomposition method decomposes features from different modalities into similarity and dissimilarity features. Due to the decoupled dissimilarity features containing both the diversity and the unique information, they show evident distribution discrepancies. Previous feature decomposition methods have overlooked the inherent contradictions in dissimilarity features, resulting in a decrease in prediction accuracy. To address this issue, a dissimilarity feature-driven decomposition network (DFDDN) for multimodal sentiment analysis is proposed. Firstly, feature extract module is used to extract and amplify features, which not only eliminate visual and audio noise but also facilitate the capture of complementary information between modalities. Secondly, different encoders are used to decouple the features, and a multimodal transformer is used to mitigate the differences in dissimilarity features. Finally, loss functions are used for optimization. Extensive experiments on two widely-used multimodal sentiment analysis datasets demonstrate the accuracy and robustness of this model, transcending SOTA performance.

With the extensive application of intelligent communication technology in smart traffic scenarios, the task of detecting pedestrians and vehicles constitutes an important technical means for road safety. In light of the high missed detection rate and slow detection speed in the foggy environment, a real-time foggy target detection method based on YOLOv8 is proposed. The model incorporates the fog removal network module into the input image to preprocess it, retains the detailed features of the original image and eliminates the obstruction of fog, and then utilizes the improved YOLOv8n for detection. On YOLOv8n, the C2f module is enhanced based on FasterNet to reduce the model parameters and size, increase the model’s computing efficiency, and the SE-ResNeXt detection head is designed to avoid the negative impacts of stacking neural network layers. Finally, knowledge distillation is employed to further enhance the detection accuracy. The proposed model is validated on the RTTS dataset and the synthetic foggy dataset. Compared with the original network, the average precision (mAP@50_95) is improved by 5.2 percentage points, and the detection frame rate reaches 170 frame/s.

In order to solve the problems of low detection accuracy, high missed detection rate and high false detection rate of bridge surface defects under the background of interference, a bridge defect detection method based on data enhancement and improved YOLOv8 is proposed. The small sample data is augmented by StyleGAN3 and depth image fusion. The SPD-Conv module is added to the YOLOv8 backbone to improve the feature extraction capability of low-resolution defects. Based on AFPN structure, AFPN_UCG structure is designed to make the network handle multi-scale information better. In C2f, RFCBAMConv and DLKA modules are introduced to construct C2f_RD module, which can transmit gradient information accurately and capture small target information more effectively. A new detection Head is designed by combining DCNv3 module with Dynamic Head, which combines three attention mechanisms of scale, space and task and uses DCNv3 dynamic adjustment to further improve the prediction performance of the model for irregular defects. Through experiments, mAP@0.5 increases by 2.4 percentage points after the data is expanded, and the accuracy rate of the improved YOLOv8 is 93.2% and mAP@0.5 is 91.3%, respectively, which are 4.2 and 4.3 percentage points higher than that of the original model, which can detect bridge defects more accurately.

For a class of two-dimensional piecewise affine system with parameters, the existence of singular cycles is obtained by adjusting the structure of the invariant manifolds of subsystems. Then, after selecting a suitable transversal in a small neighborhood of the singular cycles, the composite Poincaré map is constructed to obtain the one-side stability of the singular cycles by the flow map generated by the subsystems, and the existence and stability of the limit cycles induced by the bifurcation of the singular cycles. Finally, two examples are given to illustrate the applications of the main results.

The existence of positive normalized solutions for a class of Schrdinger equations with a perturbation for the mass supercritical case (that is, p+p²/N<q<p^*) is discussed. Firstly, by constructing an appropriate auxiliary function, it is proved that the energy functional possesses a mountain-pass geometrical structure on the constraint space. Then, the existence of a mountain pass solution with positive energy is established when h_W^1,p is sufficiently small.

The high cost of collecting covariate information is the main reason why large cohort studies or follow-up studies are halted, and Case-Cohort Design is a biased sampling mechanism to solve this problem, which has been extensively studied in survival events. However, multivariate recurrent events are also very common in biomedical and public health research, which often require long-term follow-up of trial subjects and can be costly. In view of this, this paper proposes a sampling scheme for the design of generalized case cohorts of multivariate recurrent events. By using a class of additive transfer models to fit the data, selecting the time-related weighting function, and useing the inverse probability weighting method, the unknown parameter estimation equation is established, which further proves the coincidence and asymptotic normality of the obtained parameter estimators. Finally, the effectiveness of the proposed method is verified by numerical simulation and case analysis.

Guizhou is one of the major corn-producing areas in the southwest of China, and accurate assessment of the change in the suitability of corn cultivation in Guizhou in the context of climate change is of great significance for the sustainable development of agriculture in Guizhou. In this paper, based on the historical (2001 to 2020) month-by-month meteorological data from 77 meteorological stations in Guizhou Province and the meteorological data under different emission scenarios (SSP126, SSP245, SSP585) in the future (2021 to 2040) of the CMIP6, a model of temperature, precipitation, and sunshine suitability for spring corn in Guizhou was constructed on the basis of the climate suitability function model, and its integrated suitability was analyzed. The results showed that under different emission scenarios, the temperature suitability of spring corn in Guizhou increased as a whole, and the precipitation suitability decreased slightly. Under the combined influence of temperature and precipitation, the areas with integrated suitability >0.80 under the SSP126, SSP245, and SSP585 scenarios increased from the central part of Guizhou to the east and west by 2.69%, 3.08%, and 4.05%, respectively. Since the fluctuation of precipitation suitability was small, the change of temperature was considered to play a dominant role in the zoning of spring corn planting.

Guilin City, with its many mountains and hills, is a typical karst topography. Due to the complex geological conditions and abundant rainfall, the area frequently experiences geological disasters. This paper selects Guilin as the study area, based on the geological and geomorphological, climatic and hydrological, and human activity factors of geohazards, and selects eight assessment indicators such as elevation, slope, faults, precipitation, water systems, normalized difference vegetation index (NDVI), roads, and land use types to establish an index system for the assessment of geohazards risk in Guilin City. Using the analytic hierarchy process (AHP) and the information model under the ArcGIS platform for data integration, the study area is assessed for geohazards risk. The results show that the proportions of extremely low, low, medium, high, and extremely high risk areas in Guilin City are 22.41%, 31.89%, 24.43%, 14.81%, and 6.45%, respectively. The extremely high-risk areas are mainly concentrated in the southwestern part of Guilin City, such as Yongfu County, Yangshuo County, and the urban area of Guilin; the extremely low-risk areas are mainly distributed in Longsheng County, Ziyuan County, Xing’an County, and other places; the rest of the areas have moderate risk levels, but exhibit a certain spatial differences. The model results indicate that precipitation may be the most important factor affecting the distribution of geohazards risk areas in Guilin City. The model assessment indicator (AUC) has a value of 0.796, which passes the accuracy test for the hazard assessment results. By integrating the natural geographical characteristics of the study area and the mechanisms of how each assessment indicator contributes to disasters, and comparing the hazard assessment results of typical regions, it is found that the high weight of assessment indicators such as precipitation and distance to faults, along with the medium and low weight distribution of other indicators, are quite reasonable. Furthermore, the weighted information quantity of each factor shows good correlation with its grading results, further validating the accuracy of this information quantity model. This study has assessed the impact of various disaster-causing factors on the level of geohazards risk in Guilin in detail and has delineated the geohazards risk areas in Guilin City, which has certain guiding significance for geohazards prevention and control, land use planning, and ecological and environmental protection in Guilin City and other related areas.

The quantification and assessment of ecosystem services provide critical information for scientific decision-making, which holds significant practical importance for the protection and enhancement of these services as well as the promotion of sustainable regional development. This study focuses on the Lijiang River Basin, uses the PLUS model to predict land use changes and simulate land use structures under natural development scenarios, urban development scenarios, cultivated land protection scenarios, and ecological protection scenarios in 2030. Furthermore, the InVEST model is utilized to evaluate water yield service, carbon storage service, and soil conservation service in the basin for 2010, 2020, and 2030. The conclusions were as follows: 1) Regarding land use dynamics from 2010 to 2020 in the Lijiang River Basin, the construction land and water area were increasing, while the cultivated land, forest land, and grassland areas decreased. In 2030, under multiple scenarios, the land use in the Lijiang River Basin showed a trend of conversion from cultivated land, forest land, and grassland to construction land. The expansion of construction land was evident under both natural development scenarios and urban development scenarios; under the ecological protection scenario, the declining trend of cultivated land, forest land, and grassland areas and the expansion rate of construction land were effectively controlled; the cultivated land protection scenario helps to reduce the conversion of cultivated land to construction land. 2) Concerning ecosystem services between 2010-2020 within the basin’s spatial distribution exhibited marked differences: both water yield and soil conservation demonstrated upward trends whereas carbon storage declined. The central and western parts of the basin were high-value areas for water production, while the northern and eastern parts were concentrated areas for high values of carbon storage and soil conservation. In 2030, under the urban development scenario, water production continued to increase; the differences in development scenarios had a significant impact on carbon sequestration services, and the ecological protection scenario could increase carbon storage; soil conservation services showed an increasing trend year by year and reached their maximum under the natural development scenario. 3) The proliferation of constructed environments correlates positively with enhanced water yield services but negatively impacts on forested regions leading to diminished carbon reserves; additionally, fluctuations observed indicate initial increases followed by subsequent decreases in soil conservation volumes due to construction land expansions alongside reductions in forests/grasslands. Exploring how different ecological-socioeconomic developmental modes affect ecosystem service alterations within the Lijiang River Basin can furnish valuable insights for informed spatial planning regarding land utilization along with comprehensive management practices.

Habitat quality indicates the capacity of regional ecosystems to supply essential services. Quantitatively assessing its spatial and temporal evolution and forecasting future changes are critical for high-quality regional development. This study focused on Fujian Province, where the InVEST model and GEO-detector were used to quantitatively analyze the spatial and temporal dynamics of habitat quality and its driving factors. Additionally, the PLUS model was used to simulate and predict the spatial and temporal distributions of land use and habitat quality in Fujian Province from 2020 to 2040. The findings revealed that: 1) Between 2000 and 2020, there had been a decline in cultivated land, forest land, and grassland in Fujian Province, with the most significant reductions occurring in the southeastern coastal and adjacent areas, where these lands had largely been converted to urban development. 2) Habitat quality in Fujian Province from 2000 to 2020 was generally good, with a slight decreasing trend in average habitat quality. The northwest of Nanping City, and the middle and upper areas of the Minjiang River had relatively high habitat quality, while the coastal areas of Zhangzhou, Xiamen, and Putian had low habitat quality. 3) The PLUS model predicted that under the cropland and ecological protection scenarios, the area of cropland, forest land, and grassland in Fujian Province will increase in the next 20 years, while water bodies, built-up land, and unutilized land showed a decreasing trend. This results in a gradual upward trend in the mean habitat quality and a further increase in the proportion of high habitat quality areas. 4) The first dominant factor in the change of habitat quality in Fujian Province was NPP (q=0.262), followed by GDP (q=0.199), and the results of the interaction detection showed that the interaction between the two factors of nature had a more prominent effect on the habitat quality in Fujian Province.

Gout is an acute inflammatory arthritis typically characterized by severe pain in the lower limbs. The current clinical treatment of gout, such as colchicine, non-Zithromax anti-inflammatory drugs, and so on, can cause serious side effects.Therefore, it is extremely urgent to explore a new type of effective treatment drug with few side effects. In recent years, traditional Chinese medicine (TCM) has played an advantageous role in the treatment of gout. P-coumaric acid, as a herbal ingredient with a wide range of pharmacological activities, is a highly promising drug for the treatment of gout, but its exact mechanism has not been clarified. In this rearch, mice with acute gouty arthritis were employed as a model to investigate the mitigating effect of p-coumaric acid on gout and to elucidate its potential mechanism. The research results indicates that p-coumaric acid can effectively alleviates acute gout pain. The proposed mechanisms are as follows: firstly, its anti-inflammatory properties are evident as it significantly reduces the production of interleukin-1β (IL-1β) in joint tissues of mice with acute gout. Secondly, p-coumaric acid binds to Transient Receptor Potential Vanilloid 1 (TRPV1), which can alleviate acute gouty pain by inhibiting TRPV1’s function and expression. To conclude, this rearch reveals the potential of p-coumaric acid for the treatment of gout and provides valuable insights into its mechanism of action, paving the way for further clinical applications.

Loss of gsdf in the medaka leads to excessive proliferation of germ cells and XY male germ cell sex reversal to oocytes. Transcriptomic analysis revealed a significant upregulation of the elavl2 gene expression in both XX and XY gsdf-deficient gonads, suggesting a putative inhibition of sex gland elavl2 expression by Gsdf signaling. Elavl2 had a high evolutionary conservation among species. ELAVL2 was expressed in various tissues after embryonic development and sexual maturation, with notably high expression in the gonads and brain, indicated its critical role in gametogenesis and brain function. Chemical in situ hybridization demonstrates predominant expression of elavl2 in primary spermatogonia and oocytes within normal testes and ovaries. Consistent with the relative expression of qPCR in gonads, elavl2 may play an important role in gamete production. this work supports and provides comparative evolutionary insights into the regulatory mechanisms underline the vertebrate sex differentiation.

The circadian clock governs intrinsic daily oscillations in organisms. Nevertheless, the circadian fluctuations of the Clock constituents in Quasipaa spinosa have yet to be characterized. The frog primarily exhibits diurnal behavior during its tadpole stage, but undergoes a transition to nocturnal activity upon reaching the frogling stage. To comprehend the shift from diurnal to nocturnal behavior in the evolutionary progression, it is crucial to acquire profound understanding of the light-dark circadian system exhibited by amphibians. In current study, we conducted a comprehensive analysis of the circadian clock regulation of the Clock gene in this frog. The Qsclock protein features a pivotal HLH domain, two PAS domains, and a PAC domain that demonstrates heightened conservation across various species. Evolutionary analysis revealed that the Clock gene could be classified into two distinct groups: group A and group B. Five key daily rhythmicity genes (Clock, Bmal1, Per2, Cry1, and RoRα) were specifically chosen to comprehensively investigate the transcriptional regulation involved in promoting growth in frogs throughout different tissues and developmental stages. The tissue distribution analyses revealed the ubiquitous expression of the Clock gene across all tissues and all development stages, with the highest levels of expression observed during the metamorphosis stage in all examined tissues. The analysis of cooperative regulation of rhythm revealed a synchronization in Clock and Bmal1 genes, characterized by peak levels observed during nighttime. Furthermore, both Per2 and Cry1 exhibited an early peak expression at the onset of the day. However, the expression patterns of Per2 and Cry1 exhibited tissue specificity with Cry1 genes displayed another transient peak of expression in muscle, brain, liver, and heart tissues during the T4 stage. This could potentially be linked to the increased levels of melatonin during nighttime or the behavioral patterns observed in experimental animals. The findings provide valuable insights into the intricate light-dark circadian system of amphibians.

FOXP2 gene of Fox family widely exists in the animal kingdom and is an important gene related to sound. Its mutation can cause human language disorder. FOXP2 gene is closely related to the language ability and learning behavior of birds and reptiles. In order to understand its evolutionary history, 66 species of birds and 30 species of reptiles were studied. Based on genomic data, the molecular evolution of FOXP2 gene in birds and reptiles was discussed by using selection pressure analysis and phylogenetic analysis. The results showed that: first, although both birds and reptiles had positive selection sites, FOXP2 gene was very conserved in reptiles and birds, and showed purifying selection on the whole; second, the FOXP2 gene of songbirds in birds did not undergo positive selection evolution, and there was no positive selection site, but the ω value was larger than that of other birds, indicating that the evolution rate of FOXP2 gene in songbirds is faster, which may be related to its outstanding singing ability; third, although FOXP2 of both birds and reptiles were under purifying selection, birds havd greater ω, indicating that birds were under relaxed selection and the FOXP2 gene of birds had a faster evolution rate than that of reptiles. The results of this study provide further theoretical basis for the study of the evolution and function of FOXP2 gene, and bring reference and enlightenment to the exploration of human language mechanism.