The malicious dissemination of hate speech via compromised accounts,automated bot networks and malware-driven social media campaigns has become a growing cybersecurity concern.Automatically detecting such content in S...The malicious dissemination of hate speech via compromised accounts,automated bot networks and malware-driven social media campaigns has become a growing cybersecurity concern.Automatically detecting such content in Spanish is challenging due to linguistic complexity and the scarcity of annotated resources.In this paper,we compare two predominant AI-based approaches for the forensic detection of malicious hate speech:(1)finetuning encoder-only models that have been trained in Spanish and(2)In-Context Learning techniques(Zero-and Few-Shot Learning)with large-scale language models.Our approach goes beyond binary classification,proposing a comprehensive,multidimensional evaluation that labels each text by:(1)type of speech,(2)recipient,(3)level of intensity(ordinal)and(4)targeted group(multi-label).Performance is evaluated using an annotated Spanish corpus,standard metrics such as precision,recall and F1-score and stability-oriented metrics to evaluate the stability of the transition from zero-shot to few-shot prompting(Zero-to-Few Shot Retention and Zero-to-Few Shot Gain)are applied.The results indicate that fine-tuned encoder-only models(notably MarIA and BETO variants)consistently deliver the strongest and most reliable performance:in our experiments their macro F1-scores lie roughly in the range of approximately 46%–66%depending on the task.Zero-shot approaches are much less stable and typically yield substantially lower performance(observed F1-scores range approximately 0%–39%),often producing invalid outputs in practice.Few-shot prompting(e.g.,Qwen 38B,Mistral 7B)generally improves stability and recall relative to pure zero-shot,bringing F1-scores into a moderate range of approximately 20%–51%but still falling short of fully fine-tuned models.These findings highlight the importance of supervised adaptation and discuss the potential of both paradigms as components in AI-powered cybersecurity and malware forensics systems designed to identify and mitigate coordinated online hate campaigns.展开更多
This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble lear...This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble learning techniques:DAGGING(DG),MULTIBOOST(MB),and ADABOOST(AB).This combination resulted in three distinct ensemble models:DG-RBFN,MB-RBFN,and AB-RBFN.Additionally,a traditional weighted method,Information Value(IV),and a benchmark machine learning(ML)model,Multilayer Perceptron Neural Network(MLP),were employed for comparison and validation.The models were developed using ten landslide conditioning factors,which included slope,aspect,elevation,curvature,land cover,geomorphology,overburden depth,lithology,distance to rivers and distance to roads.These factors were instrumental in predicting the output variable,which was the probability of landslide occurrence.Statistical analysis of the models’performance indicated that the DG-RBFN model,with an Area Under ROC Curve(AUC)of 0.931,outperformed the other models.The AB-RBFN model achieved an AUC of 0.929,the MB-RBFN model had an AUC of 0.913,and the MLP model recorded an AUC of 0.926.These results suggest that the advanced ensemble ML model DG-RBFN was more accurate than traditional statistical model,single MLP model,and other ensemble models in preparing trustworthy landslide susceptibility maps,thereby enhancing land use planning and decision-making.展开更多
As a key node of modern transportation network,the informationization management of road tunnels is crucial to ensure the operation safety and traffic efficiency.However,the existing tunnel vehicle modeling methods ge...As a key node of modern transportation network,the informationization management of road tunnels is crucial to ensure the operation safety and traffic efficiency.However,the existing tunnel vehicle modeling methods generally have problems such as insufficient 3D scene description capability and low dynamic update efficiency,which are difficult to meet the demand of real-time accurate management.For this reason,this paper proposes a vehicle twin modeling method for road tunnels.This approach starts from the actual management needs,and supports multi-level dynamic modeling from vehicle type,size to color by constructing a vehicle model library that can be flexibly invoked;at the same time,semantic constraint rules with geometric layout,behavioral attributes,and spatial relationships are designed to ensure that the virtual model matches with the real model with a high degree of similarity;ultimately,the prototype system is constructed and the case region is selected for the case study,and the dynamic vehicle status in the tunnel is realized by integrating real-time monitoring data with semantic constraints for precise virtual-real mapping.Finally,the prototype system is constructed and case experiments are conducted in selected case areas,which are combined with real-time monitoring data to realize dynamic updating and three-dimensional visualization of vehicle states in tunnels.The experiments show that the proposed method can run smoothly with an average rendering efficiency of 17.70 ms while guaranteeing the modeling accuracy(composite similarity of 0.867),which significantly improves the real-time and intuitive tunnel management.The research results provide reliable technical support for intelligent operation and emergency response of road tunnels,and offer new ideas for digital twin modeling of complex scenes.展开更多
The Electro–Hydrostatic Actuator(EHA)is applied to drive the control surface in flightcontrol system of more electric aircraft.In EHA,the Oil-Immersed Motor Pump(OMP)serves asthe core as a power assembly.However,the ...The Electro–Hydrostatic Actuator(EHA)is applied to drive the control surface in flightcontrol system of more electric aircraft.In EHA,the Oil-Immersed Motor Pump(OMP)serves asthe core as a power assembly.However,the compact integration of the OMP presents challenges inefficiently dissipating internal heat,leading to a performance degradation of the EHA due to ele-vated temperatures.Therefore,accurately modeling and predicting the internal thermal dynamicsof the OMP hold considerable significance for monitoring the operational condition of the EHA.In view of this,a modeling method considering cumulative thermal coupling was hereby proposed.Based on the proposed method,the thermal models of the motor and the pump were established,taking into account heat accumulation and transfer.Taking the leakage oil as the heat couplingpoint between the motor and the pump,the dynamic thermal coupling model of the OMP wasdeveloped,with the thermal characteristics of the oil considered.Additionally,the comparativeexperiments were conducted to illustrate the efficiency of the proposed model.The experimentalresults demonstrate that the proposed dynamic thermal coupling model accurately captured thethermal behavior of OMP,outperforming the static thermal parameter model.Overall,thisadvancement is crucial for effectively monitoring the health of EHA and ensuring flight safety.展开更多
As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigat...As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigation of degradation mechanisms.However,dynamic operating conditions,cell-to-cell inconsistencies,and limited availability of labeled data have posed significant challenges to accurate and robust prognosis and diagnosis.Herein,we introduce a time-series-decomposition-based ensembled lightweight learning model(TELL-Me),which employs a synergistic dual-module framework to facilitate accurate and reliable forecasting.The feature module formulates features with physical implications and sheds light on battery aging mechanisms,while the gradient module monitors capacity degradation rates and captures aging trend.TELL-Me achieves high accuracy in end-of-life prediction using minimal historical data from a single battery without requiring offline training dataset,and demonstrates impressive generality and robustness across various operating conditions and battery types.Additionally,by correlating feature contributions with degradation mechanisms across different datasets,TELL-Me is endowed with the diagnostic ability that not only enhances prediction reliability but also provides critical insights into the design and optimization of next-generation batteries.展开更多
Driven by rapid technological advancements and economic growth,mineral extraction and metal refining have increased dramatically,generating huge volumes of tailings and mine waste(TMWs).Investigating the morphological...Driven by rapid technological advancements and economic growth,mineral extraction and metal refining have increased dramatically,generating huge volumes of tailings and mine waste(TMWs).Investigating the morphological fractions of heavy metals and metalloids(HMMs)in TMWs is key to evaluating their leaching potential into the environment;however,traditional experiments are time-consuming and labor-intensive.In this study,10 machine learning(ML)algorithms were used and compared for rapidly predicting the morphological fractions of HMMs in TMWs.A dataset comprising 2376 data points was used,with mineral composition,elemental properties,and total concentration used as inputs and concentration of morphological fraction used as output.After grid search optimization,the extra tree model performed the best,achieving coefficient of determination(R2)of 0.946 and 0.942 on the validation and test sets,respectively.Electronegativity was found to have the greatest impact on the morphological fraction.The models’performance was enhanced by applying an ensemble method to the top three optimal ML models,including gradient boosting decision tree,extra trees and categorical boosting.Overall,the proposed framework can accurately predict the concentrations of different morphological fractions of HMMs in TMWs.This approach can minimize detection time,aid in the safe management and recovery of TMWs.展开更多
Pipeline Inspection Gauge(pig)is an important equipment for oil and gas pipelines during different stages of their operations to perform functions such as dewatering,cleaning,and inspection.Owing to the hyperelasticit...Pipeline Inspection Gauge(pig)is an important equipment for oil and gas pipelines during different stages of their operations to perform functions such as dewatering,cleaning,and inspection.Owing to the hyperelasticity,time and temperature-dependent material behaviour of the sealing disc attached on the pig,the contact between the pig and the pipeline expresses complex behaviour,leading to an uncertainty in the prediction of the pig's frictional force.Knowing the deformation of the sealing discs well is essential and can be highly meaningful for predicting the pig motion,as well as reducing the pigging risks.In this study,the geometrical deformation of the sealing discs with different sizes are investigated through experiments and numerical simulations.The effects of the four nondimensionalized parameters(interference,thickness per pipeline inner diameter,and clamping ratio)of the sealing discs on the deformation behaviour were observed and discussed,and an improved mathematical model for predicting the geometrical deformation of the sealing discs was proposed and verified.With the auxiliary angleαadded in the improved mathematical model,the relative error declines to 1.87%and 3.18%respectively for predicting deformation of the sealing discs in size of 2-inch and 40-inch pig.The results of this study can help better understand the frictional force of a pig with sealing discs,as well as its motion.展开更多
The Dst index has been commonly used to measure the geomagnetic effectiveness of magnetic storm events for several decades.Based on Burton’s empirical Dst model and the global magneto-hydrodynamic(MHD)simulation of E...The Dst index has been commonly used to measure the geomagnetic effectiveness of magnetic storm events for several decades.Based on Burton’s empirical Dst model and the global magneto-hydrodynamic(MHD)simulation of Earth’s magnetosphere,here we proposed a semi-empirical model to forecast the Dst index during geomagnetic storms.In this model,the ring current contribution to the Dst index is derived from Burton’s model,while the contributions from other current systems are obtained from the global MHD simulation.In order to verify the model accuracy,a number of recent magnetic storm events are tested and the simulated Dst index is compared with the observation through the correlation coefficient(CC),prediction efficiency(PE),root mean square error(RMSE)and central root mean square error(CRMSE).The results indicate that,in the context of moderate and intense geomagnetic storm events,the semi-empirical model performs well in global MHD simulations,showing relatively higher CC and PE,and lower RMSE and CRMSE compared to those from the empirical model.Compared with the physics-based ring current models,this model inherits the advantage of fast processing from the empirical model,and easy implementation in a global MHD model of Earth’s magnetosphere.Therefore,it is suitable for the Dst estimation under a context of a global MHD simulation.展开更多
Federated semi-supervised learning(FSSL)faces two major challenges:the scarcity of labeled data across clients and the non-independent and identically distributed(Non-IID)nature of data among clients.To address these ...Federated semi-supervised learning(FSSL)faces two major challenges:the scarcity of labeled data across clients and the non-independent and identically distributed(Non-IID)nature of data among clients.To address these issues,we propose diffusion model-based data synthesis aided FSSL(DDSA-FSSL),a novel approach that leverages diffusion model(DM)to generate synthetic data,thereby bridging the gap between heterogeneous local data distributions and the global data distribution.In the proposed DDSA-FSSL,each client addresses the scarcity of labeled data by utilizing a federated learningtrained classifier to perform pseudo labeling for unlabeled data.The DM is then collaboratively trained using both labeled and precision-optimized pseudolabeled data,enabling clients to generate synthetic samples for classes that are absent in their labeled datasets.As a result,the disparity between local and global distributions is reduced and clients can create enriched synthetic datasets that better align with the global data distribution.Extensive experiments on various datasets and Non-IID scenarios demonstrate the effectiveness of DDSA-FSSL,achieving significant performance improvements,such as increasing accuracy from 38.46%to 52.14%on CIFAR-10 datasets with 10%labeled data.展开更多
In recent decades,large ensemble simulation(LENS)or super-large ensemble simulation(SLENS)experiments with climate models,including the simulation of both the historical and future climate,have been increasingly explo...In recent decades,large ensemble simulation(LENS)or super-large ensemble simulation(SLENS)experiments with climate models,including the simulation of both the historical and future climate,have been increasingly exploited in the fields of climate change,climate variability,climate projection,and beyond.This paper provides an overview of LENS in climate systems.It delves into its definition,initialization,significance,and scientific concerns.Additionally,its development history and relevant theories,methods,and primary fields of application are also reviewed.Conclusions obtained from single-model LENS can be more robust compared with those from ensemble simulations with smaller numbers of members.The interactions among model biases,forced responses,and internal variabilities,which serve as the added value in LENS,are highlighted.Finally,we put forward the future trajectory of LENS with climate or Earth system models(ESMs).Super-large ensemble simulation,high-resolution LENS,LENS employing ESMs,and combining LENS with artificial intelligence,will greatly promote the study of climate and related applications.展开更多
Based on the hindcasts from five subseasonal-to-seasonal(S2S)models participating in the S2S Prediction Project,this study evaluates the performance of the multimodel ensemble(MME)approach in predicting the subseasona...Based on the hindcasts from five subseasonal-to-seasonal(S2S)models participating in the S2S Prediction Project,this study evaluates the performance of the multimodel ensemble(MME)approach in predicting the subseasonal precipitation anomalies during summer in China and reveals the contributions of possible driving factors.The results suggest that while single-model ensembles(SMEs)exhibit constrained predictive skills within a limited forecast lead time of three pentads,the MME illustrates an enhanced predictive skill at a lead time of up to four pentads,and even six pentads,in southern China.Based on both deterministic and probabilistic verification metrics,the MME consistently outperforms SMEs,with a more evident advantage observed in probabilistic forecasting.The superior performance of the MME is primarily attributable to the increase in ensemble size,and the enhanced model diversity is also a contributing factor.The reliability of probabilistic skill is largely improved due to the increase in ensemble members,while the resolution term does not exhibit consistent improvement.Furthermore,the Madden–Julian Oscillation(MJO)is revealed as the primary driving factor for the successful prediction of summer precipitation in China using the MME.The improvement by the MME is not solely attributable to the enhancement in the inherent predictive capacity of the MJO itself,but derives from its capability in capturing the more realistic relationship between the MJO and subseasonal precipitation anomalies in China.This study establishes a scientific foundation for acknowledging the advantageous predictive capability of the MME approach in subseasonal predictions of summer precipitation in China,and sheds light on further improving S2S predictions.展开更多
To reveal the deterioration mechanism of coal-rock assemblages under chemical corrosion and dynamic loading,chemical corrosion and dynamic impact experiments were conducted.Under different chemical corrosion condition...To reveal the deterioration mechanism of coal-rock assemblages under chemical corrosion and dynamic loading,chemical corrosion and dynamic impact experiments were conducted.Under different chemical corrosion conditions,the weakening characteristics,observable characteristics,softening characteristics of the dynamic parameters,dynamic failure characteristics,dynamic failure forms and dynamic microscopic characteristics were analyzed.Under each corrosion condition,the dynamic elastic modulus,dynamic deformation modulus and dynamic peak intensity tended to decrease with immersing time.The dynamic elastic modulus,dynamic deformation modulus and dynamic peak intensity exhibited an inverted U-shaped trend.Under dynamic impact,the failure process of acidly corroded samples can be divided into the following stages:the initial stage,elastic energy accumulation stage,local failure of coal and secondary rock crack expansion stage,coal fragment ejection stage,rock spalling stage and complete instability stage.Under dynamic impact,failure modes exist:coal crushing failure,rock fragmenting failure,rock splitting failure and full splitting failure.After impact failure,sample fragments are distributed in powder,granular,cone and block forms.Based on Zhu-Wang-Tang nonlinear viscoelastic properties,a model considering chemical corrosion and impact damage was proposed.The combined effects of chemical and impact-induced damage on the dynamic mechanical properties of coal-rock assemblages were systematically analyzed.展开更多
Rice is one of the most important staple crops globally.Rice plant diseases can severely reduce crop yields and,in extreme cases,lead to total production loss.Early diagnosis enables timely intervention,mitigates dise...Rice is one of the most important staple crops globally.Rice plant diseases can severely reduce crop yields and,in extreme cases,lead to total production loss.Early diagnosis enables timely intervention,mitigates disease severity,supports effective treatment strategies,and reduces reliance on excessive pesticide use.Traditional machine learning approaches have been applied for automated rice disease diagnosis;however,these methods depend heavily on manual image preprocessing and handcrafted feature extraction,which are labor-intensive and time-consuming and often require domain expertise.Recently,end-to-end deep learning(DL) models have been introduced for this task,but they often lack robustness and generalizability across diverse datasets.To address these limitations,we propose a novel end-toend training framework for convolutional neural network(CNN) and attention-based model ensembles(E2ETCA).This framework integrates features from two state-of-the-art(SOTA) CNN models,Inception V3 and DenseNet-201,and an attention-based vision transformer(ViT) model.The fused features are passed through an additional fully connected layer with softmax activation for final classification.The entire process is trained end-to-end,enhancing its suitability for realworld deployment.Furthermore,we extract and analyze the learned features using a support vector machine(SVM),a traditional machine learning classifier,to provide comparative insights.We evaluate the proposed E2ETCA framework on three publicly available datasets,the Mendeley Rice Leaf Disease Image Samples dataset,the Kaggle Rice Diseases Image dataset,the Bangladesh Rice Research Institute dataset,and a combined version of all three.Using standard evaluation metrics(accuracy,precision,recall,and F1-score),our framework demonstrates superior performance compared to existing SOTA methods in rice disease diagnosis,with potential applicability to other agricultural disease detection tasks.展开更多
Workpiece rotational grinding is widely used in the ultra-precision machining of hard and brittle semiconductor materials,including single-crystal silicon,silicon carbide,and gallium arsenide.Surface roughness and sub...Workpiece rotational grinding is widely used in the ultra-precision machining of hard and brittle semiconductor materials,including single-crystal silicon,silicon carbide,and gallium arsenide.Surface roughness and subsurface damage depth(SDD)are crucial indicators for evaluating the surface quality of these materials after grinding.Existing prediction models lack general applicability and do not accurately account for the complex material behavior under grinding conditions.This paper introduces novel models for predicting both surface roughness and SDD in hard and brittle semiconductor materials.The surface roughness model uniquely incorporates the material’s elastic recovery properties,revealing the significant impact of these properties on prediction accuracy.The SDD model is distinguished by its analysis of the interactions between abrasive grits and the workpiece,as well as the mechanisms governing stress-induced damage evolution.The surface roughness model and SDD model both establish a stable relationship with the grit depth of cut(GDC).Additionally,we have developed an analytical relationship between the GDC and grinding process parameters.This,in turn,enables the establishment of an analytical framework for predicting surface roughness and SDD based on grinding process parameters,which cannot be achieved by previous models.The models were validated through systematic experiments on three different semiconductor materials,demonstrating excellent agreement with experimental data,with prediction errors of 6.3%for surface roughness and6.9%for SDD.Additionally,this study identifies variations in elastic recovery and material plasticity as critical factors influencing surface roughness and SDD across different materials.These findings significantly advance the accuracy of predictive models and broaden their applicability for grinding hard and brittle semiconductor materials.展开更多
Semi-crystalline polymer laser powder bed fusion(L-PBF)has recently attracted increasing interest due to its potential for fabricating complex geometry.However,a more comprehensive understanding of the underlying phys...Semi-crystalline polymer laser powder bed fusion(L-PBF)has recently attracted increasing interest due to its potential for fabricating complex geometry.However,a more comprehensive understanding of the underlying physics during L-PBF is required to better control the properties of the final part.This work proposed a multi-layer numerical model to study the temperature and phase evolution during the polyamide-12(PA12)L-PBF process.The Descend and Parallel Chord methods were introduced to improve the convergence of the non-linear thermal solver.The level-set-based mesh adaptation strategy,governed by multi-physical fields,was applied to alleviate the calculation and accurately track the phase evolution.The processing simulation on the dog-bone model revealed that preheating temperature significantly influences the crystallization behavior.Finally,the multi-layer simulation demonstrated that such a developed numerical model can be used to study the phase transformation during powder layer updating and the cyclic laser sintering phenomena.Moreover,the numerical study suggested that crystallization occurs slowly during the L-PBF process.展开更多
This study adopts a communication studies perspective,employing Lasswell’s“5W”model as the theoretical framework and integrating the“6W”elements of translation communication to systematically examine the translat...This study adopts a communication studies perspective,employing Lasswell’s“5W”model as the theoretical framework and integrating the“6W”elements of translation communication to systematically examine the translation and dissemination model of the English version of Mo Yan’s novel“Life and Death Are Wearing Me Out,”translated by Howard Goldblatt.By analyzing key elements such as communication content,translator strategies,dissemination channels,and reader reception,the study elucidates the intrinsic mechanisms behind the work’s successful overseas dissemination.The findings reveal that effective communication stems from the synergistic interplay of four dimensions:the intercultural adaptability of the work itself,the translator’s agential mediation,diversified dissemination channels,and reader-oriented translation strategies.This research provides theoretical and practical insights for promoting Chinese literature globally,demonstrating that successful intercultural communication requires an integrated approach encompassing the work,translator,channels,and reader research.展开更多
Reptile fauna should be considered a conservation objective,especially in respect of the impacts of climate change on their distribution and range’s dynamics.Investigating the environmental drivers of reptile species...Reptile fauna should be considered a conservation objective,especially in respect of the impacts of climate change on their distribution and range’s dynamics.Investigating the environmental drivers of reptile species richness and identifying their suitable habitats is a fundamental prerequisite to setting efficient long-term conservation measures.This study focused on geographical patterns and estimations of species richness for herpetofauna widely spread Z.vivipara,N.natrix,V.berus,A.colchica,and protected in Latvia C.austriaca,E.orbicularis,L.agilis inhabiting northern(model territory Latvia)and southern(model territory Ukraine)part of their European range.The ultimate goal was to designate a conservation network that will meet long-term goals for survival of the target species in the context of climate change.We used stacked species distribution models for creating maps depicting the distribution of species richness under current and future(by 2050)climates for marginal reptilepopulations.Using cluster analysis,we showed that this herpeto-complex can be divided into“widespread species”and“forest species”.For all forest species we predicted a climate-driven reduction in their distribution range both North(Latvia)and South(Ukraine).The most vulnerable populations of“forest species”tend to be located in the South of their range,as a consequence of northward shifts by 2050.By 2050 the greatest reduction in range is predicted for currently widely spread Z.vivipara(by 1.4 times)and V.berus(by 2.2 times).In terms of designing an effective protected-area network,these results permit to identify priority conservation areas where the full ensemble of selected reptile species can be found,and confirms the relevance of abioticmulti-factor GIS-modelling for achieving this goal.展开更多
基金the research project LaTe4PoliticES(PID2022-138099OB-I00)funded by MCIN/AEI/10.13039/501100011033 and the European Fund for Regional Development(ERDF)-a way to make Europe.Tomás Bernal-Beltrán is supported by University of Murcia through the predoctoral programme.
文摘The malicious dissemination of hate speech via compromised accounts,automated bot networks and malware-driven social media campaigns has become a growing cybersecurity concern.Automatically detecting such content in Spanish is challenging due to linguistic complexity and the scarcity of annotated resources.In this paper,we compare two predominant AI-based approaches for the forensic detection of malicious hate speech:(1)finetuning encoder-only models that have been trained in Spanish and(2)In-Context Learning techniques(Zero-and Few-Shot Learning)with large-scale language models.Our approach goes beyond binary classification,proposing a comprehensive,multidimensional evaluation that labels each text by:(1)type of speech,(2)recipient,(3)level of intensity(ordinal)and(4)targeted group(multi-label).Performance is evaluated using an annotated Spanish corpus,standard metrics such as precision,recall and F1-score and stability-oriented metrics to evaluate the stability of the transition from zero-shot to few-shot prompting(Zero-to-Few Shot Retention and Zero-to-Few Shot Gain)are applied.The results indicate that fine-tuned encoder-only models(notably MarIA and BETO variants)consistently deliver the strongest and most reliable performance:in our experiments their macro F1-scores lie roughly in the range of approximately 46%–66%depending on the task.Zero-shot approaches are much less stable and typically yield substantially lower performance(observed F1-scores range approximately 0%–39%),often producing invalid outputs in practice.Few-shot prompting(e.g.,Qwen 38B,Mistral 7B)generally improves stability and recall relative to pure zero-shot,bringing F1-scores into a moderate range of approximately 20%–51%but still falling short of fully fine-tuned models.These findings highlight the importance of supervised adaptation and discuss the potential of both paradigms as components in AI-powered cybersecurity and malware forensics systems designed to identify and mitigate coordinated online hate campaigns.
基金the University of Transport Technology under the project entitled“Application of Machine Learning Algorithms in Landslide Susceptibility Mapping in Mountainous Areas”with grant number DTTD2022-16.
文摘This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble learning techniques:DAGGING(DG),MULTIBOOST(MB),and ADABOOST(AB).This combination resulted in three distinct ensemble models:DG-RBFN,MB-RBFN,and AB-RBFN.Additionally,a traditional weighted method,Information Value(IV),and a benchmark machine learning(ML)model,Multilayer Perceptron Neural Network(MLP),were employed for comparison and validation.The models were developed using ten landslide conditioning factors,which included slope,aspect,elevation,curvature,land cover,geomorphology,overburden depth,lithology,distance to rivers and distance to roads.These factors were instrumental in predicting the output variable,which was the probability of landslide occurrence.Statistical analysis of the models’performance indicated that the DG-RBFN model,with an Area Under ROC Curve(AUC)of 0.931,outperformed the other models.The AB-RBFN model achieved an AUC of 0.929,the MB-RBFN model had an AUC of 0.913,and the MLP model recorded an AUC of 0.926.These results suggest that the advanced ensemble ML model DG-RBFN was more accurate than traditional statistical model,single MLP model,and other ensemble models in preparing trustworthy landslide susceptibility maps,thereby enhancing land use planning and decision-making.
基金National Natural Science Foundation of China(Nos.42301473,42271424,42171397)Chinese Postdoctoral Innovation Talents Support Program(No.BX20230299)+2 种基金China Postdoctoral Science Foundation(No.2023M742884)Natural Science Foundation of Sichuan Province(Nos.24NSFSC2264,2025ZNSFSC0322)Key Research and Development Project of Sichuan Province(No.24ZDYF0633).
文摘As a key node of modern transportation network,the informationization management of road tunnels is crucial to ensure the operation safety and traffic efficiency.However,the existing tunnel vehicle modeling methods generally have problems such as insufficient 3D scene description capability and low dynamic update efficiency,which are difficult to meet the demand of real-time accurate management.For this reason,this paper proposes a vehicle twin modeling method for road tunnels.This approach starts from the actual management needs,and supports multi-level dynamic modeling from vehicle type,size to color by constructing a vehicle model library that can be flexibly invoked;at the same time,semantic constraint rules with geometric layout,behavioral attributes,and spatial relationships are designed to ensure that the virtual model matches with the real model with a high degree of similarity;ultimately,the prototype system is constructed and the case region is selected for the case study,and the dynamic vehicle status in the tunnel is realized by integrating real-time monitoring data with semantic constraints for precise virtual-real mapping.Finally,the prototype system is constructed and case experiments are conducted in selected case areas,which are combined with real-time monitoring data to realize dynamic updating and three-dimensional visualization of vehicle states in tunnels.The experiments show that the proposed method can run smoothly with an average rendering efficiency of 17.70 ms while guaranteeing the modeling accuracy(composite similarity of 0.867),which significantly improves the real-time and intuitive tunnel management.The research results provide reliable technical support for intelligent operation and emergency response of road tunnels,and offer new ideas for digital twin modeling of complex scenes.
基金supported by the National Key R&D Program of China(No.2021YFB2011300)the National Natural Science Foundation of China(Nos.52275044,U2233212)。
文摘The Electro–Hydrostatic Actuator(EHA)is applied to drive the control surface in flightcontrol system of more electric aircraft.In EHA,the Oil-Immersed Motor Pump(OMP)serves asthe core as a power assembly.However,the compact integration of the OMP presents challenges inefficiently dissipating internal heat,leading to a performance degradation of the EHA due to ele-vated temperatures.Therefore,accurately modeling and predicting the internal thermal dynamicsof the OMP hold considerable significance for monitoring the operational condition of the EHA.In view of this,a modeling method considering cumulative thermal coupling was hereby proposed.Based on the proposed method,the thermal models of the motor and the pump were established,taking into account heat accumulation and transfer.Taking the leakage oil as the heat couplingpoint between the motor and the pump,the dynamic thermal coupling model of the OMP wasdeveloped,with the thermal characteristics of the oil considered.Additionally,the comparativeexperiments were conducted to illustrate the efficiency of the proposed model.The experimentalresults demonstrate that the proposed dynamic thermal coupling model accurately captured thethermal behavior of OMP,outperforming the static thermal parameter model.Overall,thisadvancement is crucial for effectively monitoring the health of EHA and ensuring flight safety.
基金supported by the National Natural Science Foundation of China(22379021 and 22479021)。
文摘As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigation of degradation mechanisms.However,dynamic operating conditions,cell-to-cell inconsistencies,and limited availability of labeled data have posed significant challenges to accurate and robust prognosis and diagnosis.Herein,we introduce a time-series-decomposition-based ensembled lightweight learning model(TELL-Me),which employs a synergistic dual-module framework to facilitate accurate and reliable forecasting.The feature module formulates features with physical implications and sheds light on battery aging mechanisms,while the gradient module monitors capacity degradation rates and captures aging trend.TELL-Me achieves high accuracy in end-of-life prediction using minimal historical data from a single battery without requiring offline training dataset,and demonstrates impressive generality and robustness across various operating conditions and battery types.Additionally,by correlating feature contributions with degradation mechanisms across different datasets,TELL-Me is endowed with the diagnostic ability that not only enhances prediction reliability but also provides critical insights into the design and optimization of next-generation batteries.
基金Project(2024JJ2074) supported by the Natural Science Foundation of Hunan Province,ChinaProject(22376221) supported by the National Natural Science Foundation of ChinaProject(2023QNRC001) supported by the Young Elite Scientists Sponsorship Program by CAST,China。
文摘Driven by rapid technological advancements and economic growth,mineral extraction and metal refining have increased dramatically,generating huge volumes of tailings and mine waste(TMWs).Investigating the morphological fractions of heavy metals and metalloids(HMMs)in TMWs is key to evaluating their leaching potential into the environment;however,traditional experiments are time-consuming and labor-intensive.In this study,10 machine learning(ML)algorithms were used and compared for rapidly predicting the morphological fractions of HMMs in TMWs.A dataset comprising 2376 data points was used,with mineral composition,elemental properties,and total concentration used as inputs and concentration of morphological fraction used as output.After grid search optimization,the extra tree model performed the best,achieving coefficient of determination(R2)of 0.946 and 0.942 on the validation and test sets,respectively.Electronegativity was found to have the greatest impact on the morphological fraction.The models’performance was enhanced by applying an ensemble method to the top three optimal ML models,including gradient boosting decision tree,extra trees and categorical boosting.Overall,the proposed framework can accurately predict the concentrations of different morphological fractions of HMMs in TMWs.This approach can minimize detection time,aid in the safe management and recovery of TMWs.
基金supported by Key Technologies Research and Development Program(Grant No.SQ2022YFC2806103)and the National Natural Science Foundation of China(Grant No.51509259).
文摘Pipeline Inspection Gauge(pig)is an important equipment for oil and gas pipelines during different stages of their operations to perform functions such as dewatering,cleaning,and inspection.Owing to the hyperelasticity,time and temperature-dependent material behaviour of the sealing disc attached on the pig,the contact between the pig and the pipeline expresses complex behaviour,leading to an uncertainty in the prediction of the pig's frictional force.Knowing the deformation of the sealing discs well is essential and can be highly meaningful for predicting the pig motion,as well as reducing the pigging risks.In this study,the geometrical deformation of the sealing discs with different sizes are investigated through experiments and numerical simulations.The effects of the four nondimensionalized parameters(interference,thickness per pipeline inner diameter,and clamping ratio)of the sealing discs on the deformation behaviour were observed and discussed,and an improved mathematical model for predicting the geometrical deformation of the sealing discs was proposed and verified.With the auxiliary angleαadded in the improved mathematical model,the relative error declines to 1.87%and 3.18%respectively for predicting deformation of the sealing discs in size of 2-inch and 40-inch pig.The results of this study can help better understand the frictional force of a pig with sealing discs,as well as its motion.
基金supported by NNSFC grants 42150101,42188105,42304189National Key R&D program of China No.2021YFA-0718600the Pandeng Program of National Space Science Center,Chinese Academy of Sciences.
文摘The Dst index has been commonly used to measure the geomagnetic effectiveness of magnetic storm events for several decades.Based on Burton’s empirical Dst model and the global magneto-hydrodynamic(MHD)simulation of Earth’s magnetosphere,here we proposed a semi-empirical model to forecast the Dst index during geomagnetic storms.In this model,the ring current contribution to the Dst index is derived from Burton’s model,while the contributions from other current systems are obtained from the global MHD simulation.In order to verify the model accuracy,a number of recent magnetic storm events are tested and the simulated Dst index is compared with the observation through the correlation coefficient(CC),prediction efficiency(PE),root mean square error(RMSE)and central root mean square error(CRMSE).The results indicate that,in the context of moderate and intense geomagnetic storm events,the semi-empirical model performs well in global MHD simulations,showing relatively higher CC and PE,and lower RMSE and CRMSE compared to those from the empirical model.Compared with the physics-based ring current models,this model inherits the advantage of fast processing from the empirical model,and easy implementation in a global MHD model of Earth’s magnetosphere.Therefore,it is suitable for the Dst estimation under a context of a global MHD simulation.
基金supported in part by NSF of China under Grant 62222111 and Grant 62431015in part by the Science and Technology Commission Foundation of Shanghai under Grant 24DP1500702.
文摘Federated semi-supervised learning(FSSL)faces two major challenges:the scarcity of labeled data across clients and the non-independent and identically distributed(Non-IID)nature of data among clients.To address these issues,we propose diffusion model-based data synthesis aided FSSL(DDSA-FSSL),a novel approach that leverages diffusion model(DM)to generate synthetic data,thereby bridging the gap between heterogeneous local data distributions and the global data distribution.In the proposed DDSA-FSSL,each client addresses the scarcity of labeled data by utilizing a federated learningtrained classifier to perform pseudo labeling for unlabeled data.The DM is then collaboratively trained using both labeled and precision-optimized pseudolabeled data,enabling clients to generate synthetic samples for classes that are absent in their labeled datasets.As a result,the disparity between local and global distributions is reduced and clients can create enriched synthetic datasets that better align with the global data distribution.Extensive experiments on various datasets and Non-IID scenarios demonstrate the effectiveness of DDSA-FSSL,achieving significant performance improvements,such as increasing accuracy from 38.46%to 52.14%on CIFAR-10 datasets with 10%labeled data.
基金This study was supported by the National Natural Science Foundation of China(Grant No.U2342228)the National Key Program for Developing Basic Sciences(Grant No.2020YFA0608902)+1 种基金the National Natural Science Foundation of China(Grant Nos.92358302,and 42242018)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDB0500303).
文摘In recent decades,large ensemble simulation(LENS)or super-large ensemble simulation(SLENS)experiments with climate models,including the simulation of both the historical and future climate,have been increasingly exploited in the fields of climate change,climate variability,climate projection,and beyond.This paper provides an overview of LENS in climate systems.It delves into its definition,initialization,significance,and scientific concerns.Additionally,its development history and relevant theories,methods,and primary fields of application are also reviewed.Conclusions obtained from single-model LENS can be more robust compared with those from ensemble simulations with smaller numbers of members.The interactions among model biases,forced responses,and internal variabilities,which serve as the added value in LENS,are highlighted.Finally,we put forward the future trajectory of LENS with climate or Earth system models(ESMs).Super-large ensemble simulation,high-resolution LENS,LENS employing ESMs,and combining LENS with artificial intelligence,will greatly promote the study of climate and related applications.
基金sponsored by the National Natural Science Foundation of China(Grant Nos.42175052 and U2442206)the Joint Research Project for Meteorological Capacity Improvement(Grant No.23NLTSQ007,23NLTSZ003)+2 种基金the Innovative Development Special Project of the China Meteorological Administration(Grant No.CXFZ2023J002)the National Key R&D Program of China(Grant No.2023YFC3007700,2024YFC3013100)the China Meteorological Administration Youth Innovation Team(Grant No.CMA2024QN06)。
文摘Based on the hindcasts from five subseasonal-to-seasonal(S2S)models participating in the S2S Prediction Project,this study evaluates the performance of the multimodel ensemble(MME)approach in predicting the subseasonal precipitation anomalies during summer in China and reveals the contributions of possible driving factors.The results suggest that while single-model ensembles(SMEs)exhibit constrained predictive skills within a limited forecast lead time of three pentads,the MME illustrates an enhanced predictive skill at a lead time of up to four pentads,and even six pentads,in southern China.Based on both deterministic and probabilistic verification metrics,the MME consistently outperforms SMEs,with a more evident advantage observed in probabilistic forecasting.The superior performance of the MME is primarily attributable to the increase in ensemble size,and the enhanced model diversity is also a contributing factor.The reliability of probabilistic skill is largely improved due to the increase in ensemble members,while the resolution term does not exhibit consistent improvement.Furthermore,the Madden–Julian Oscillation(MJO)is revealed as the primary driving factor for the successful prediction of summer precipitation in China using the MME.The improvement by the MME is not solely attributable to the enhancement in the inherent predictive capacity of the MJO itself,but derives from its capability in capturing the more realistic relationship between the MJO and subseasonal precipitation anomalies in China.This study establishes a scientific foundation for acknowledging the advantageous predictive capability of the MME approach in subseasonal predictions of summer precipitation in China,and sheds light on further improving S2S predictions.
基金supported by the National Natural Science Foundation of China(Nos.52034009 and 52174093)the Fundamental Research Funds for the Central Universities(Nos.2024ZKPYNY01,2023ZKPYNY03,and 2023YQTD02).
文摘To reveal the deterioration mechanism of coal-rock assemblages under chemical corrosion and dynamic loading,chemical corrosion and dynamic impact experiments were conducted.Under different chemical corrosion conditions,the weakening characteristics,observable characteristics,softening characteristics of the dynamic parameters,dynamic failure characteristics,dynamic failure forms and dynamic microscopic characteristics were analyzed.Under each corrosion condition,the dynamic elastic modulus,dynamic deformation modulus and dynamic peak intensity tended to decrease with immersing time.The dynamic elastic modulus,dynamic deformation modulus and dynamic peak intensity exhibited an inverted U-shaped trend.Under dynamic impact,the failure process of acidly corroded samples can be divided into the following stages:the initial stage,elastic energy accumulation stage,local failure of coal and secondary rock crack expansion stage,coal fragment ejection stage,rock spalling stage and complete instability stage.Under dynamic impact,failure modes exist:coal crushing failure,rock fragmenting failure,rock splitting failure and full splitting failure.After impact failure,sample fragments are distributed in powder,granular,cone and block forms.Based on Zhu-Wang-Tang nonlinear viscoelastic properties,a model considering chemical corrosion and impact damage was proposed.The combined effects of chemical and impact-induced damage on the dynamic mechanical properties of coal-rock assemblages were systematically analyzed.
基金the Begum Rokeya University,Rangpur,and the United Arab Emirates University,UAE for partially supporting this work。
文摘Rice is one of the most important staple crops globally.Rice plant diseases can severely reduce crop yields and,in extreme cases,lead to total production loss.Early diagnosis enables timely intervention,mitigates disease severity,supports effective treatment strategies,and reduces reliance on excessive pesticide use.Traditional machine learning approaches have been applied for automated rice disease diagnosis;however,these methods depend heavily on manual image preprocessing and handcrafted feature extraction,which are labor-intensive and time-consuming and often require domain expertise.Recently,end-to-end deep learning(DL) models have been introduced for this task,but they often lack robustness and generalizability across diverse datasets.To address these limitations,we propose a novel end-toend training framework for convolutional neural network(CNN) and attention-based model ensembles(E2ETCA).This framework integrates features from two state-of-the-art(SOTA) CNN models,Inception V3 and DenseNet-201,and an attention-based vision transformer(ViT) model.The fused features are passed through an additional fully connected layer with softmax activation for final classification.The entire process is trained end-to-end,enhancing its suitability for realworld deployment.Furthermore,we extract and analyze the learned features using a support vector machine(SVM),a traditional machine learning classifier,to provide comparative insights.We evaluate the proposed E2ETCA framework on three publicly available datasets,the Mendeley Rice Leaf Disease Image Samples dataset,the Kaggle Rice Diseases Image dataset,the Bangladesh Rice Research Institute dataset,and a combined version of all three.Using standard evaluation metrics(accuracy,precision,recall,and F1-score),our framework demonstrates superior performance compared to existing SOTA methods in rice disease diagnosis,with potential applicability to other agricultural disease detection tasks.
基金supported by the National Key Research and Development Program of China(2022YFB3605902)the National Natural Science Foundation of China(52375411,52293402)。
文摘Workpiece rotational grinding is widely used in the ultra-precision machining of hard and brittle semiconductor materials,including single-crystal silicon,silicon carbide,and gallium arsenide.Surface roughness and subsurface damage depth(SDD)are crucial indicators for evaluating the surface quality of these materials after grinding.Existing prediction models lack general applicability and do not accurately account for the complex material behavior under grinding conditions.This paper introduces novel models for predicting both surface roughness and SDD in hard and brittle semiconductor materials.The surface roughness model uniquely incorporates the material’s elastic recovery properties,revealing the significant impact of these properties on prediction accuracy.The SDD model is distinguished by its analysis of the interactions between abrasive grits and the workpiece,as well as the mechanisms governing stress-induced damage evolution.The surface roughness model and SDD model both establish a stable relationship with the grit depth of cut(GDC).Additionally,we have developed an analytical relationship between the GDC and grinding process parameters.This,in turn,enables the establishment of an analytical framework for predicting surface roughness and SDD based on grinding process parameters,which cannot be achieved by previous models.The models were validated through systematic experiments on three different semiconductor materials,demonstrating excellent agreement with experimental data,with prediction errors of 6.3%for surface roughness and6.9%for SDD.Additionally,this study identifies variations in elastic recovery and material plasticity as critical factors influencing surface roughness and SDD across different materials.These findings significantly advance the accuracy of predictive models and broaden their applicability for grinding hard and brittle semiconductor materials.
文摘Semi-crystalline polymer laser powder bed fusion(L-PBF)has recently attracted increasing interest due to its potential for fabricating complex geometry.However,a more comprehensive understanding of the underlying physics during L-PBF is required to better control the properties of the final part.This work proposed a multi-layer numerical model to study the temperature and phase evolution during the polyamide-12(PA12)L-PBF process.The Descend and Parallel Chord methods were introduced to improve the convergence of the non-linear thermal solver.The level-set-based mesh adaptation strategy,governed by multi-physical fields,was applied to alleviate the calculation and accurately track the phase evolution.The processing simulation on the dog-bone model revealed that preheating temperature significantly influences the crystallization behavior.Finally,the multi-layer simulation demonstrated that such a developed numerical model can be used to study the phase transformation during powder layer updating and the cyclic laser sintering phenomena.Moreover,the numerical study suggested that crystallization occurs slowly during the L-PBF process.
基金2024 Jiangsu Normal University Postgraduate Research and Practice Innovation Program(Project No.:2024XKT1750)。
文摘This study adopts a communication studies perspective,employing Lasswell’s“5W”model as the theoretical framework and integrating the“6W”elements of translation communication to systematically examine the translation and dissemination model of the English version of Mo Yan’s novel“Life and Death Are Wearing Me Out,”translated by Howard Goldblatt.By analyzing key elements such as communication content,translator strategies,dissemination channels,and reader reception,the study elucidates the intrinsic mechanisms behind the work’s successful overseas dissemination.The findings reveal that effective communication stems from the synergistic interplay of four dimensions:the intercultural adaptability of the work itself,the translator’s agential mediation,diversified dissemination channels,and reader-oriented translation strategies.This research provides theoretical and practical insights for promoting Chinese literature globally,demonstrating that successful intercultural communication requires an integrated approach encompassing the work,translator,channels,and reader research.
文摘Reptile fauna should be considered a conservation objective,especially in respect of the impacts of climate change on their distribution and range’s dynamics.Investigating the environmental drivers of reptile species richness and identifying their suitable habitats is a fundamental prerequisite to setting efficient long-term conservation measures.This study focused on geographical patterns and estimations of species richness for herpetofauna widely spread Z.vivipara,N.natrix,V.berus,A.colchica,and protected in Latvia C.austriaca,E.orbicularis,L.agilis inhabiting northern(model territory Latvia)and southern(model territory Ukraine)part of their European range.The ultimate goal was to designate a conservation network that will meet long-term goals for survival of the target species in the context of climate change.We used stacked species distribution models for creating maps depicting the distribution of species richness under current and future(by 2050)climates for marginal reptilepopulations.Using cluster analysis,we showed that this herpeto-complex can be divided into“widespread species”and“forest species”.For all forest species we predicted a climate-driven reduction in their distribution range both North(Latvia)and South(Ukraine).The most vulnerable populations of“forest species”tend to be located in the South of their range,as a consequence of northward shifts by 2050.By 2050 the greatest reduction in range is predicted for currently widely spread Z.vivipara(by 1.4 times)and V.berus(by 2.2 times).In terms of designing an effective protected-area network,these results permit to identify priority conservation areas where the full ensemble of selected reptile species can be found,and confirms the relevance of abioticmulti-factor GIS-modelling for achieving this goal.
