With the rapid growth of the Internet and social media, information is widely disseminated in multimodal forms, such as text and images, where discriminatory content can manifest in various ways. Discrimination detect...With the rapid growth of the Internet and social media, information is widely disseminated in multimodal forms, such as text and images, where discriminatory content can manifest in various ways. Discrimination detection techniques for multilingual and multimodal data can identify potential discriminatory behavior and help foster a more equitable and inclusive cyberspace. However, existing methods often struggle in complex contexts and multilingual environments. To address these challenges, this paper proposes an innovative detection method, using image and multilingual text encoders to separately extract features from different modalities. It continuously updates a historical feature memory bank, aggregates the Top-K most similar samples, and utilizes a Gated Recurrent Unit (GRU) to integrate current and historical features, generating enhanced feature representations with stronger semantic expressiveness to improve the model’s ability to capture discriminatory signals. Experimental results demonstrate that the proposed method exhibits superior discriminative power and detection accuracy in multilingual and multimodal contexts, offering a reliable and effective solution for identifying discriminatory content.展开更多
Taking the rural low-income population of Zhejiang Province as its subject, this paper examines how to build a sustainable income-growth mechanism and identify feasible implementation paths within the context of the c...Taking the rural low-income population of Zhejiang Province as its subject, this paper examines how to build a sustainable income-growth mechanism and identify feasible implementation paths within the context of the common prosperity strategy. The research identifies key obstacles to income expansion, including an undiversified industrial structure, insufficient human capital, and a lack of robust social protection. These call for systemic solutions featuring institutional innovation, resource consolidation, and capability enhancement. Building on Zhejiang's experience as a common prosperity demonstration zone, the article constructs an integrated framework centered on four pillars: industrial empowerment, education upgrading, social security reinforcement, and digital coordination. It further offers concrete policy proposals involving the cultivation of localized industries, vocational skill training, enhanced safety nets, and the adoption of digital tools. The study thus offers both theoretical insights and practical paradigms for tackling the challenge of raising incomes in low-income rural areas.展开更多
With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performan...With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.展开更多
In the image fusion field,fusing infrared images(IRIs)and visible images(VIs)excelled is a key area.The differences between IRIs and VIs make it challenging to fuse both types into a high-quality image.Accordingly,eff...In the image fusion field,fusing infrared images(IRIs)and visible images(VIs)excelled is a key area.The differences between IRIs and VIs make it challenging to fuse both types into a high-quality image.Accordingly,efficiently combining the advantages of both images while overcoming their shortcomings is necessary.To handle this challenge,we developed an end-to-end IRI andVI fusionmethod based on frequency decomposition and enhancement.By applying concepts from frequency domain analysis,we used the layering mechanism to better capture the salient thermal targets from the IRIs and the rich textural information from the VIs,respectively,significantly boosting the image fusion quality and effectiveness.In addition,the backbone network combined Restormer Blocks and Dense Blocks;Restormer blocks utilize global attention to extract shallow features.Meanwhile,Dense Blocks ensure the integration between shallow and deep features,thereby avoiding the loss of shallow attributes.Extensive experiments on TNO and MSRS datasets demonstrated that the suggested method achieved state-of-the-art(SOTA)performance in various metrics:Entropy(EN),Mutual Information(MI),Standard Deviation(SD),The Structural Similarity Index Measure(SSIM),Fusion quality(Qabf),MI of the pixel(FMI_(pixel)),and modified Visual Information Fidelity(VIF_(m)).展开更多
Autumn rain in western China(ARWC) is a unique and significant precipitation phenomenon that occurs during the summer-to-winter transition of the atmospheric circulation. Using the fifth generation of global climate a...Autumn rain in western China(ARWC) is a unique and significant precipitation phenomenon that occurs during the summer-to-winter transition of the atmospheric circulation. Using the fifth generation of global climate and weather reanalysis data from the European Centre for Medium-Range Weather Forecasts and CN05.1 grid precipitation data, this study examined the anomalous characteristics and mechanisms of ARWC by combining the synergistic effect of the westerly jet and meridional wind. Over the past 60 years, ARWC has exhibited significant interdecadal and interannual variations, as well as a north-south seesaw pattern. The westerly jet index(meridional wind index) exhibited a negative(positive) correlation with precipitation in the northern autumn rain zone(NARZ), and a positive(negative) correlation with precipitation in the southern autumn rain zone(SARZ). The coupling of a weak meridional southerly wind with a southward westerly jet and a strong meridional southerly wind with a northward westerly jet are the two primary modes that synergistically influence the ARWC. These synergistic effects cause significant atmospheric changes throughout the troposphere. The contrasting circulation structure, temperature advection, vertical motion, and water vapor flux contributed to the opposite precipitation anomalies observed in the NARZ and SARZ. A new comprehensive index that reflects the coupled synergistic effect is proposed to characterize the anomalous changes in ARWC. This study improves the understanding of the anomalous characteristics and mechanisms of ARWC.展开更多
This study addresses the risk of privacy leakage during the transmission and sharing of multimodal data in smart grid substations by proposing a three-tier privacy-preserving architecture based on asynchronous federat...This study addresses the risk of privacy leakage during the transmission and sharing of multimodal data in smart grid substations by proposing a three-tier privacy-preserving architecture based on asynchronous federated learning.The framework integrates blockchain technology,the InterPlanetary File System(IPFS)for distributed storage,and a dynamic differential privacy mechanism to achieve collaborative security across the storage,service,and federated coordination layers.It accommodates both multimodal data classification and object detection tasks,enabling the identification and localization of key targets and abnormal behaviors in substation scenarios while ensuring privacy protection.This effectively mitigates the single-point failures and model leakage issues inherent in centralized architectures.A dynamically adjustable differential privacy mechanism is introduced to allocate privacy budgets according to client contribution levels and upload frequencies,achieving a personalized balance between model performance and privacy protection.Multi-dimensional experimental evaluations,including classification accuracy,F1-score,encryption latency,and aggregation latency,verify the security and efficiency of the proposed architecture.The improved CNN model achieves 72.34%accuracy and an F1-score of 0.72 in object detection and classification tasks on infrared surveillance imagery,effectively identifying typical risk events such as not wearing safety helmets and unauthorized intrusion,while maintaining an aggregation latency of only 1.58 s and a query latency of 80.79 ms.Compared with traditional static differential privacy and centralized approaches,the proposed method demonstrates significant advantages in accuracy,latency,and security,providing a new technical paradigm for efficient,secure data sharing,object detection,and privacy preservation in smart grid substations.展开更多
In recent years,the research on superconductivity in one-dimensional(1D)materials has been attracting increasing attention due to its potential applications in low-dimensional nanodevices.However,the critical temperat...In recent years,the research on superconductivity in one-dimensional(1D)materials has been attracting increasing attention due to its potential applications in low-dimensional nanodevices.However,the critical temperature(T_(c))of 1D superconductors is low.In this work,we theoretically investigate the possible high T_(c) superconductivity of(5,5)carbon nanotube(CNT).The pristine(5,5)CNT is a Dirac semimetal and can be modulated into a semiconductor by full hydrogenation.Interestingly,by further hole doping,it can be regulated into a metallic state with the sp^(3)-hybridized σ electrons metalized,and a giant Kohn anomaly appears in the optical phonons.The two factors together enhance the electron–phonon coupling,and lead to high-T_(c) superconductivity.When the hole doping concentration of hydrogenated-(5,5)CNT is 2.5 hole/cell,the calculated T_(c) is 82.3 K,exceeding the boiling point of liquid nitrogen.Therefore,the predicted hole-doped hydrogenated-(5,5)CNT provides a new platform for 1D high-T_(c) superconductivity and may have potential applications in 1D nanodevices.展开更多
A scheme is proposed based on a Mach-Zehnder interferometer with high phase sensitivity,utilizing a two-mode squeezed coherent state,generated by four-wave mixing,as input.The phase sensitivity of this scheme easily s...A scheme is proposed based on a Mach-Zehnder interferometer with high phase sensitivity,utilizing a two-mode squeezed coherent state,generated by four-wave mixing,as input.The phase sensitivity of this scheme easily surpasses the Heisenberg limit when intensity difference detection is applied.Under phase-matching conditions,the quantum Cramér-Rao bound significantly exceeds the Heisenberg limit.Additionally,the scheme exhibits robustness against photon loss.When compared with the modified SU(1,1)interferometer with two coherent state inputs,this approach demonstrates superior measurement sensitivity,evaluated through various detection methods and the quantum Cramér-Rao bound.This work holds potential applications in quantum metrology.展开更多
Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that man...Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that manipulate model behavior through malicious instructions.Following Kitchenham’s guidelines,this systematic review synthesizes 128 peer-reviewed studies from 2022 to 2025 to provide a unified understanding of this rapidly evolving threat landscape.Our findings reveal a swift progression from simple direct injections to sophisticated multimodal attacks,achieving over 90%success rates against unprotected systems.In response,defense mechanisms show varying effectiveness:input preprocessing achieves 60%–80%detection rates and advanced architectural defenses demonstrate up to 95%protection against known patterns,though significant gaps persist against novel attack vectors.We identified 37 distinct defense approaches across three categories,but standardized evaluation frameworks remain limited.Our analysis attributes these vulnerabilities to fundamental LLM architectural limitations,such as the inability to distinguish instructions from data and attention mechanism vulnerabilities.This highlights critical research directions such as formal verification methods,standardized evaluation protocols,and architectural innovations for inherently secure LLM designs.展开更多
Accurately predicting the synthesizability of inorganic crystal materials serves as a pivotal tool for the efficient screening of viable candidates,substantially reducing the costs associated with extensive experiment...Accurately predicting the synthesizability of inorganic crystal materials serves as a pivotal tool for the efficient screening of viable candidates,substantially reducing the costs associated with extensive experimental trial-and-error processes.However,existing methods,limited by static structural descriptors such as chemical composition and lattice parameters,fail to account for atomic vibrations,which may introduce spurious correlations and undermine predictive reliability.Here,we propose a deep learning model termed integrating graph and dynamical stability(IGDS)for predicting the synthesizability of inorganic crystals.IGDS employs graph representation learning to construct crystal graphs that precisely capture the static structures of crystals and integrates phonon spectral features extracted from pre-trained machine learning interatomic potentials to represent their dynamic properties.Our model exhibits outstanding performance in predicting the synthesizability of low-energy unsynthesizable crystals across 41 material systems,achieving precision and recall values of 0.916/0.863 for ternary compounds.By capturing both static structural descriptors and dynamic features,IGDS provides a physics-informed method for predicting the synthesizability of inorganic crystals.This approach bridges the gap between theoretical design concepts and their practical implementation,thereby streamlining the development cycle of new materials and enhancing overall research efficiency.展开更多
With the large-scale deployment of the Internet ofThings(IoT)devices,their weak securitymechanisms make them prime targets for malware attacks.Attackers often use Domain Generation Algorithm(DGA)to generate random dom...With the large-scale deployment of the Internet ofThings(IoT)devices,their weak securitymechanisms make them prime targets for malware attacks.Attackers often use Domain Generation Algorithm(DGA)to generate random domain names,hiding the real IP of Command and Control(C&C)servers to build botnets.Due to the randomness and dynamics of DGA,traditional methods struggle to detect them accurately,increasing the difficulty of network defense.This paper proposes a lightweight DGA detection model based on knowledge distillation for resource-constrained IoT environments.Specifically,a teacher model combining CharacterBERT,a bidirectional long short-term memory(BiLSTM)network,and attention mechanism(ATT)is constructed:it extracts character-level semantic features viaCharacterBERT,captures sequence dependencieswith the BiLSTM,and integrates theATT for key feature weighting,formingmulti-granularity feature fusion.An improved knowledge distillation approach transfers the teacher model’s learned knowledge to the simplified DistilBERT student model.Experimental results show the teacher model achieves 98.68%detection accuracy.The student modelmaintains slightly improved accuracy while significantly compressing parameters to approximately 38.4%of the teacher model’s scale,greatly reducing computational overhead for IoT deployment.展开更多
To address the issues of insufficient and imbalanced data samples in proton exchange membrane fuel cell(PEMFC)performance degradation prediction,this study proposes a data augmentation-based model to predict PEMFC per...To address the issues of insufficient and imbalanced data samples in proton exchange membrane fuel cell(PEMFC)performance degradation prediction,this study proposes a data augmentation-based model to predict PEMFC performance degradation.Firstly,an improved generative adversarial network(IGAN)with adaptive gradient penalty coefficient is proposed to address the problems of excessively fast gradient descent and insufficient diversity of generated samples.Then,the IGANis used to generate datawith a distribution analogous to real data,therebymitigating the insufficiency and imbalance of original PEMFC samples and providing the predictionmodel with training data rich in feature information.Finally,a convolutional neural network-bidirectional long short-termmemory(CNN-BiLSTM)model is adopted to predict PEMFC performance degradation.Experimental results show that the data generated by the proposed IGAN exhibits higher quality than that generated by the original GAN,and can fully characterize and enrich the original data’s features.Using the augmented data,the prediction accuracy of the CNN-BiLSTM model is significantly improved,rendering it applicable to tasks of predicting PEMFC performance degradation.展开更多
Rapid regional population shifts and spatial polarization have heightened pressure on cultivated land—a critical resource demanding urgent attention amid ongoing urban-rural transition.This study selects Jiangsu prov...Rapid regional population shifts and spatial polarization have heightened pressure on cultivated land—a critical resource demanding urgent attention amid ongoing urban-rural transition.This study selects Jiangsu province,a national leader in both economic and agricultural development,as a case area to construct a multidimensional framework for assessing the recessive morphological characteristics of multifunctional cultivated land use.We examine temporal dynamics,spatial heterogeneity,and propose an integrated zoning strategy based on empirical analysis.The results reveal that:(1)The recessive morphology index shows a consistent upward trend,with structural breaks in 2007 and 2013,and a spatial shift from“higher in the east and lower in the west”to“higher in the south and lower in the north.”(2)Coordination among sub-dimensions of the index has steadily improved.(3)The index is expected to continue rising in the next decade,though at a slower pace.(4)To promote coordinated multidimensional land-use development,we recommend a policy framework that reinforces existing strengths,addresses weaknesses,and adapts zoning schemes to current spatial conditions.This research offers new insights into multifunctional cultivated land systems and underscores their role in enhancing human well-being,securing food supply,and supporting sustainable urban-rural integration.展开更多
The Intrusion Detection System(IDS)is a security mechanism developed to observe network traffic and recognize suspicious or malicious activities.Clustering algorithms are often incorporated into IDS;however,convention...The Intrusion Detection System(IDS)is a security mechanism developed to observe network traffic and recognize suspicious or malicious activities.Clustering algorithms are often incorporated into IDS;however,conventional clustering-based methods face notable drawbacks,including poor scalability in handling high-dimensional datasets and a strong dependence of outcomes on initial conditions.To overcome the performance limitations of existing methods,this study proposes a novel quantum-inspired clustering algorithm that relies on a similarity coefficient-based quantum genetic algorithm(SC-QGA)and an improved quantum artificial bee colony algorithm hybrid K-means(IQABC-K).First,the SC-QGA algorithmis constructed based on quantum computing and integrates similarity coefficient theory to strengthen genetic diversity and feature extraction capabilities.For the subsequent clustering phase,the process based on the IQABC-K algorithm is enhanced with the core improvement of adaptive rotation gate and movement exploitation strategies to balance the exploration capabilities of global search and the exploitation capabilities of local search.Simultaneously,the acceleration of convergence toward the global optimum and a reduction in computational complexity are facilitated by means of the global optimum bootstrap strategy and a linear population reduction strategy.Through experimental evaluation with multiple algorithms and diverse performance metrics,the proposed algorithm confirms reliable accuracy on three datasets:KDD CUP99,NSL_KDD,and UNSW_NB15,achieving accuracy of 98.57%,98.81%,and 98.32%,respectively.These results affirm its potential as an effective solution for practical clustering applications.展开更多
OBJECTIVE:To study the morphological basis of the role of Siqi(cold as winter,cool as autumn,warm as spring,hot as summer),Wuwei(five flavors:sweet,pungent,salty,sour,and bitter),and Guijing(meridian tropism) through ...OBJECTIVE:To study the morphological basis of the role of Siqi(cold as winter,cool as autumn,warm as spring,hot as summer),Wuwei(five flavors:sweet,pungent,salty,sour,and bitter),and Guijing(meridian tropism) through the use of information integration.METHODS:A14C-2-deoxy-glucose autoradiography method was adopted to determine the overall impact of treatment with 39 herbs on functions of various tissues and organs.Data was measured at 4hs after a single dose and following the last treatment of repeated doses for a week.Least-squares estimation was used and fitted for each herb regression effect of organs and tissues after singleand repeated treatment.The slope of the regression line represented the cumulative trend of the effect of the herbs(β),and the standard deviation of the slope(Sβ) was compared with those of the untreated animals(t 'test).All significantly cumulative effect trends were applied with an artificial neural network,which integrated the relationship among Siqi,Wuwei,and Guijing with tissues and organs.RESULTS:There is a certain relationship among the Siqi,Wuwei,Guijing and the anatomy of organs and tissues,but the different scores indicate that influence of Siqi,Wuwei,Guijing to anatomy of organs and tissues was a nonlinear state.CONCLUSION:Results demonstrated that the effects of Siqi,Wuwei,and Guijing have a morphological basis,and each concept was associated with multiple anatomical structures.展开更多
This paper has designed a unified storage model of Automatic Weather Station Quality Information Inquiries and Feedback( AWSQIF),and has developed an AWSQIF automatic management system,which implements the storage of ...This paper has designed a unified storage model of Automatic Weather Station Quality Information Inquiries and Feedback( AWSQIF),and has developed an AWSQIF automatic management system,which implements the storage of AWSQIF automatically in national quality information two-way feedback process. Based on the unified storage,AWSQIF data have important application in real-time data management,data quality assessment and quality control effect analysis. The results show that the data quality has been improved significantly,and the data assessment has achieved online operating. In addition,AWSQIF can assist the improvement of the quality control effect.展开更多
Aiming at the integrative management and comprehensive applications of large-scale 3D geospatial information covering the full 3D space of a city, this paper briefly introduces the design and implementation of a full ...Aiming at the integrative management and comprehensive applications of large-scale 3D geospatial information covering the full 3D space of a city, this paper briefly introduces the design and implementation of a full 3D GIS platform： GeoScope, which provides a professional solution for the massive full three-dimensional geospatial data integration, management, analysis, visualization, and applications. GeoScope is characterized by： （1） extendible software architecture based on the hierarchical message bus, facilitates multimodal integrative applications of 2D GIS and 3D GIS; （2） unified 3D city models, support multiscale se- mantic representation of outdoor ＆ indoor and aboveground ＆ underground 3D objects; （3） high-efficient 3D geospatial database engine, supports integrated management of massive 3D geospatial data for real-time applications; and （4） high-performance visualization engine exploiting the massively parallel computation architecture of modem GPUs, supports real-time realistic rendering of large-scale complicated 3D geospatial environments. The successful pilot application of GeoScope is also illustrated with the 3D city models of 8494 knl2 of the whole Wuhan City, the largest city in middle China.展开更多
Analyzed the main problems which were found in current conditions and prob- lems of informationization in coal enterprises.It clarified how to achieve informationization in coal mine and put forward a general configur...Analyzed the main problems which were found in current conditions and prob- lems of informationization in coal enterprises.It clarified how to achieve informationization in coal mine and put forward a general configuration of informationization construction in which informationization in coal enterprises was divided into two parts: informationization of safety production and informationization of management.Planned a platform of inte- grated management of informationization in coal enterprises.Ultimately,it has brought forward that an overall integrated digital mine is the way to achieve the goal of informa- tionization in coal enterprises,which can promote the application of automation,digitaliza- tion,networking,informaitionization to intellectualization.At the same time,the competi- tiveness of enterprises can be improved entirely,and new type of coal industry can be supported by information technology.展开更多
Marine geographic information system （MGIS） has great ability to deal with the spatio-temporal problems and has potential superiority when it is applied to oceanography. Using the feature extraction of oceanic pheno...Marine geographic information system （MGIS） has great ability to deal with the spatio-temporal problems and has potential superiority when it is applied to oceanography. Using the feature extraction of oceanic phenomena as a case study, the functions of the MGIS are analyzed, and thus the position of MGIS in the oceanography is defined. Comparing the requirement of MGIS with that of the traditional GIS which has been developed in the terrestrial applications in the past four decades, the frame for the functions of MGIS is constructed. According to the established MGIS, some key technologies are discussed in detail with emphasis on the specialities which can distinguish the MGIS from the traditional GIS.展开更多
During the 12^(th)Five-Year Plan period,the environmental protection information work made remarkable achievements,and the construction of nationwide air quality monitoring network was completed; the trinitarian monit...During the 12^(th)Five-Year Plan period,the environmental protection information work made remarkable achievements,and the construction of nationwide air quality monitoring network was completed; the trinitarian monitoring system was established,and environmental emergency command ability enhanced significantly. However,the issues of " information island" and " application silo" are still outstanding,and the application and promotion of new technologies make slow progress; the capacity of information system to support environmental protection business is insufficient,and there is no capacity for decision analysis. In the 13^(th)Five-Year Plan period,it is needed to strengthen the application and promotion of new technologies such as cloud computing,big data,Internet of Things,and spatial information,mainly construct and improve " Cloud Computing Based Environmental Protection( CCBEP) ", " big data platform of ecological environment",and then realize the core objective of improving environmental quality.展开更多
基金funded by the Open Foundation of Key Laboratory of Cyberspace Security,Ministry of Education[KLCS20240210].
文摘With the rapid growth of the Internet and social media, information is widely disseminated in multimodal forms, such as text and images, where discriminatory content can manifest in various ways. Discrimination detection techniques for multilingual and multimodal data can identify potential discriminatory behavior and help foster a more equitable and inclusive cyberspace. However, existing methods often struggle in complex contexts and multilingual environments. To address these challenges, this paper proposes an innovative detection method, using image and multilingual text encoders to separately extract features from different modalities. It continuously updates a historical feature memory bank, aggregates the Top-K most similar samples, and utilizes a Gated Recurrent Unit (GRU) to integrate current and historical features, generating enhanced feature representations with stronger semantic expressiveness to improve the model’s ability to capture discriminatory signals. Experimental results demonstrate that the proposed method exhibits superior discriminative power and detection accuracy in multilingual and multimodal contexts, offering a reliable and effective solution for identifying discriminatory content.
文摘Taking the rural low-income population of Zhejiang Province as its subject, this paper examines how to build a sustainable income-growth mechanism and identify feasible implementation paths within the context of the common prosperity strategy. The research identifies key obstacles to income expansion, including an undiversified industrial structure, insufficient human capital, and a lack of robust social protection. These call for systemic solutions featuring institutional innovation, resource consolidation, and capability enhancement. Building on Zhejiang's experience as a common prosperity demonstration zone, the article constructs an integrated framework centered on four pillars: industrial empowerment, education upgrading, social security reinforcement, and digital coordination. It further offers concrete policy proposals involving the cultivation of localized industries, vocational skill training, enhanced safety nets, and the adoption of digital tools. The study thus offers both theoretical insights and practical paradigms for tackling the challenge of raising incomes in low-income rural areas.
基金supported by the Natural Science Foundation of China No.62362008the Major Scientific and Technological Special Project of Guizhou Province([2024]014).
文摘With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.
基金funded by Anhui Province University Key Science and Technology Project(2024AH053415)Anhui Province University Major Science and Technology Project(2024AH040229)+3 种基金Talent Research Initiation Fund Project of Tongling University(2024tlxyrc019)Tongling University School-Level Scientific Research Project(2024tlxyptZD07)TheUniversity Synergy Innovation Programof Anhui Province(GXXT-2023-050)Tongling City Science and Technology Major Special Project(Unveiling and Commanding Model)(200401JB004).
文摘In the image fusion field,fusing infrared images(IRIs)and visible images(VIs)excelled is a key area.The differences between IRIs and VIs make it challenging to fuse both types into a high-quality image.Accordingly,efficiently combining the advantages of both images while overcoming their shortcomings is necessary.To handle this challenge,we developed an end-to-end IRI andVI fusionmethod based on frequency decomposition and enhancement.By applying concepts from frequency domain analysis,we used the layering mechanism to better capture the salient thermal targets from the IRIs and the rich textural information from the VIs,respectively,significantly boosting the image fusion quality and effectiveness.In addition,the backbone network combined Restormer Blocks and Dense Blocks;Restormer blocks utilize global attention to extract shallow features.Meanwhile,Dense Blocks ensure the integration between shallow and deep features,thereby avoiding the loss of shallow attributes.Extensive experiments on TNO and MSRS datasets demonstrated that the suggested method achieved state-of-the-art(SOTA)performance in various metrics:Entropy(EN),Mutual Information(MI),Standard Deviation(SD),The Structural Similarity Index Measure(SSIM),Fusion quality(Qabf),MI of the pixel(FMI_(pixel)),and modified Visual Information Fidelity(VIF_(m)).
基金National Natural Science Foundation of China(42030611)Second Tibetan Plateau Scientific Expedition and Research (STEP) Program (2019QZKK0103)。
文摘Autumn rain in western China(ARWC) is a unique and significant precipitation phenomenon that occurs during the summer-to-winter transition of the atmospheric circulation. Using the fifth generation of global climate and weather reanalysis data from the European Centre for Medium-Range Weather Forecasts and CN05.1 grid precipitation data, this study examined the anomalous characteristics and mechanisms of ARWC by combining the synergistic effect of the westerly jet and meridional wind. Over the past 60 years, ARWC has exhibited significant interdecadal and interannual variations, as well as a north-south seesaw pattern. The westerly jet index(meridional wind index) exhibited a negative(positive) correlation with precipitation in the northern autumn rain zone(NARZ), and a positive(negative) correlation with precipitation in the southern autumn rain zone(SARZ). The coupling of a weak meridional southerly wind with a southward westerly jet and a strong meridional southerly wind with a northward westerly jet are the two primary modes that synergistically influence the ARWC. These synergistic effects cause significant atmospheric changes throughout the troposphere. The contrasting circulation structure, temperature advection, vertical motion, and water vapor flux contributed to the opposite precipitation anomalies observed in the NARZ and SARZ. A new comprehensive index that reflects the coupled synergistic effect is proposed to characterize the anomalous changes in ARWC. This study improves the understanding of the anomalous characteristics and mechanisms of ARWC.
基金funded by the National Natural Science Foundation of China,grant number 61605004the Fundamental Research Funds for the Central Universities,grant number FRF-TP-19-016A2Guizhou Power Grid Co.,Ltd.2024 first batch of services(2024-2026 technology R&D services for science and technology projects(in addition to national and SGCC key projects)),grant number 060100KC23100012。
文摘This study addresses the risk of privacy leakage during the transmission and sharing of multimodal data in smart grid substations by proposing a three-tier privacy-preserving architecture based on asynchronous federated learning.The framework integrates blockchain technology,the InterPlanetary File System(IPFS)for distributed storage,and a dynamic differential privacy mechanism to achieve collaborative security across the storage,service,and federated coordination layers.It accommodates both multimodal data classification and object detection tasks,enabling the identification and localization of key targets and abnormal behaviors in substation scenarios while ensuring privacy protection.This effectively mitigates the single-point failures and model leakage issues inherent in centralized architectures.A dynamically adjustable differential privacy mechanism is introduced to allocate privacy budgets according to client contribution levels and upload frequencies,achieving a personalized balance between model performance and privacy protection.Multi-dimensional experimental evaluations,including classification accuracy,F1-score,encryption latency,and aggregation latency,verify the security and efficiency of the proposed architecture.The improved CNN model achieves 72.34%accuracy and an F1-score of 0.72 in object detection and classification tasks on infrared surveillance imagery,effectively identifying typical risk events such as not wearing safety helmets and unauthorized intrusion,while maintaining an aggregation latency of only 1.58 s and a query latency of 80.79 ms.Compared with traditional static differential privacy and centralized approaches,the proposed method demonstrates significant advantages in accuracy,latency,and security,providing a new technical paradigm for efficient,secure data sharing,object detection,and privacy preservation in smart grid substations.
基金supported by the National Natural Science Foundation of China (Grant Nos.12074213 and 11574108)the Major Basic Program of Natural Science Foundation of Shandong Province (Grant No.ZR2021ZD01)the Natural Science Foundation of Shandong Province (Grant No.ZR2023MA082)。
文摘In recent years,the research on superconductivity in one-dimensional(1D)materials has been attracting increasing attention due to its potential applications in low-dimensional nanodevices.However,the critical temperature(T_(c))of 1D superconductors is low.In this work,we theoretically investigate the possible high T_(c) superconductivity of(5,5)carbon nanotube(CNT).The pristine(5,5)CNT is a Dirac semimetal and can be modulated into a semiconductor by full hydrogenation.Interestingly,by further hole doping,it can be regulated into a metallic state with the sp^(3)-hybridized σ electrons metalized,and a giant Kohn anomaly appears in the optical phonons.The two factors together enhance the electron–phonon coupling,and lead to high-T_(c) superconductivity.When the hole doping concentration of hydrogenated-(5,5)CNT is 2.5 hole/cell,the calculated T_(c) is 82.3 K,exceeding the boiling point of liquid nitrogen.Therefore,the predicted hole-doped hydrogenated-(5,5)CNT provides a new platform for 1D high-T_(c) superconductivity and may have potential applications in 1D nanodevices.
基金supported by the National Natural Science Foundation of China(Grant Nos.12104190,12104189,12204312)the Natural Science Foundation of Jiangsu Province(Grant No.BK20210874)+2 种基金General project of Natural Science Research in Colleges And Universities of Jiangsu Province(Grant No.20KJB140008)the Jiangxi Provincial Natural Science Foundation(Grant Nos.20224BAB211014 and 20232BAB201042)Key Laboratory of Tian Qin Project(Sun Yat-sen University)。
文摘A scheme is proposed based on a Mach-Zehnder interferometer with high phase sensitivity,utilizing a two-mode squeezed coherent state,generated by four-wave mixing,as input.The phase sensitivity of this scheme easily surpasses the Heisenberg limit when intensity difference detection is applied.Under phase-matching conditions,the quantum Cramér-Rao bound significantly exceeds the Heisenberg limit.Additionally,the scheme exhibits robustness against photon loss.When compared with the modified SU(1,1)interferometer with two coherent state inputs,this approach demonstrates superior measurement sensitivity,evaluated through various detection methods and the quantum Cramér-Rao bound.This work holds potential applications in quantum metrology.
基金supported by 2023 Higher Education Scientific Research Planning Project of China Society of Higher Education(No.23PG0408)2023 Philosophy and Social Science Research Programs in Jiangsu Province(No.2023SJSZ0993)+2 种基金Nantong Science and Technology Project(No.JC2023070)Key Project of Jiangsu Province Education Science 14th Five-Year Plan(Grant No.B-b/2024/02/41)the Open Fund of Advanced Cryptography and System Security Key Laboratory of Sichuan Province(Grant No.SKLACSS-202407).
文摘Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that manipulate model behavior through malicious instructions.Following Kitchenham’s guidelines,this systematic review synthesizes 128 peer-reviewed studies from 2022 to 2025 to provide a unified understanding of this rapidly evolving threat landscape.Our findings reveal a swift progression from simple direct injections to sophisticated multimodal attacks,achieving over 90%success rates against unprotected systems.In response,defense mechanisms show varying effectiveness:input preprocessing achieves 60%–80%detection rates and advanced architectural defenses demonstrate up to 95%protection against known patterns,though significant gaps persist against novel attack vectors.We identified 37 distinct defense approaches across three categories,but standardized evaluation frameworks remain limited.Our analysis attributes these vulnerabilities to fundamental LLM architectural limitations,such as the inability to distinguish instructions from data and attention mechanism vulnerabilities.This highlights critical research directions such as formal verification methods,standardized evaluation protocols,and architectural innovations for inherently secure LLM designs.
文摘Accurately predicting the synthesizability of inorganic crystal materials serves as a pivotal tool for the efficient screening of viable candidates,substantially reducing the costs associated with extensive experimental trial-and-error processes.However,existing methods,limited by static structural descriptors such as chemical composition and lattice parameters,fail to account for atomic vibrations,which may introduce spurious correlations and undermine predictive reliability.Here,we propose a deep learning model termed integrating graph and dynamical stability(IGDS)for predicting the synthesizability of inorganic crystals.IGDS employs graph representation learning to construct crystal graphs that precisely capture the static structures of crystals and integrates phonon spectral features extracted from pre-trained machine learning interatomic potentials to represent their dynamic properties.Our model exhibits outstanding performance in predicting the synthesizability of low-energy unsynthesizable crystals across 41 material systems,achieving precision and recall values of 0.916/0.863 for ternary compounds.By capturing both static structural descriptors and dynamic features,IGDS provides a physics-informed method for predicting the synthesizability of inorganic crystals.This approach bridges the gap between theoretical design concepts and their practical implementation,thereby streamlining the development cycle of new materials and enhancing overall research efficiency.
基金supported by the following projects:National Natural Science Foundation of China(62461041)Natural Science Foundation of Jiangxi Province China(20242BAB25068).
文摘With the large-scale deployment of the Internet ofThings(IoT)devices,their weak securitymechanisms make them prime targets for malware attacks.Attackers often use Domain Generation Algorithm(DGA)to generate random domain names,hiding the real IP of Command and Control(C&C)servers to build botnets.Due to the randomness and dynamics of DGA,traditional methods struggle to detect them accurately,increasing the difficulty of network defense.This paper proposes a lightweight DGA detection model based on knowledge distillation for resource-constrained IoT environments.Specifically,a teacher model combining CharacterBERT,a bidirectional long short-term memory(BiLSTM)network,and attention mechanism(ATT)is constructed:it extracts character-level semantic features viaCharacterBERT,captures sequence dependencieswith the BiLSTM,and integrates theATT for key feature weighting,formingmulti-granularity feature fusion.An improved knowledge distillation approach transfers the teacher model’s learned knowledge to the simplified DistilBERT student model.Experimental results show the teacher model achieves 98.68%detection accuracy.The student modelmaintains slightly improved accuracy while significantly compressing parameters to approximately 38.4%of the teacher model’s scale,greatly reducing computational overhead for IoT deployment.
基金supported by the Jiangsu Engineering Research Center of the Key Technology for Intelligent Manufacturing Equipment and the Suqian Key Laboratory of Intelligent Manufacturing(Grant No.M202108).
文摘To address the issues of insufficient and imbalanced data samples in proton exchange membrane fuel cell(PEMFC)performance degradation prediction,this study proposes a data augmentation-based model to predict PEMFC performance degradation.Firstly,an improved generative adversarial network(IGAN)with adaptive gradient penalty coefficient is proposed to address the problems of excessively fast gradient descent and insufficient diversity of generated samples.Then,the IGANis used to generate datawith a distribution analogous to real data,therebymitigating the insufficiency and imbalance of original PEMFC samples and providing the predictionmodel with training data rich in feature information.Finally,a convolutional neural network-bidirectional long short-termmemory(CNN-BiLSTM)model is adopted to predict PEMFC performance degradation.Experimental results show that the data generated by the proposed IGAN exhibits higher quality than that generated by the original GAN,and can fully characterize and enrich the original data’s features.Using the augmented data,the prediction accuracy of the CNN-BiLSTM model is significantly improved,rendering it applicable to tasks of predicting PEMFC performance degradation.
基金National Natural Science Foundation of China,No.42101252。
文摘Rapid regional population shifts and spatial polarization have heightened pressure on cultivated land—a critical resource demanding urgent attention amid ongoing urban-rural transition.This study selects Jiangsu province,a national leader in both economic and agricultural development,as a case area to construct a multidimensional framework for assessing the recessive morphological characteristics of multifunctional cultivated land use.We examine temporal dynamics,spatial heterogeneity,and propose an integrated zoning strategy based on empirical analysis.The results reveal that:(1)The recessive morphology index shows a consistent upward trend,with structural breaks in 2007 and 2013,and a spatial shift from“higher in the east and lower in the west”to“higher in the south and lower in the north.”(2)Coordination among sub-dimensions of the index has steadily improved.(3)The index is expected to continue rising in the next decade,though at a slower pace.(4)To promote coordinated multidimensional land-use development,we recommend a policy framework that reinforces existing strengths,addresses weaknesses,and adapts zoning schemes to current spatial conditions.This research offers new insights into multifunctional cultivated land systems and underscores their role in enhancing human well-being,securing food supply,and supporting sustainable urban-rural integration.
基金supported by the NSFC(Grant Nos.62176273,62271070,62441212)The Open Foundation of State Key Laboratory of Networking and Switching Technology(Beijing University of Posts and Telecommunications)under Grant SKLNST-2024-1-062025Major Project of the Natural Science Foundation of Inner Mongolia(2025ZD008).
文摘The Intrusion Detection System(IDS)is a security mechanism developed to observe network traffic and recognize suspicious or malicious activities.Clustering algorithms are often incorporated into IDS;however,conventional clustering-based methods face notable drawbacks,including poor scalability in handling high-dimensional datasets and a strong dependence of outcomes on initial conditions.To overcome the performance limitations of existing methods,this study proposes a novel quantum-inspired clustering algorithm that relies on a similarity coefficient-based quantum genetic algorithm(SC-QGA)and an improved quantum artificial bee colony algorithm hybrid K-means(IQABC-K).First,the SC-QGA algorithmis constructed based on quantum computing and integrates similarity coefficient theory to strengthen genetic diversity and feature extraction capabilities.For the subsequent clustering phase,the process based on the IQABC-K algorithm is enhanced with the core improvement of adaptive rotation gate and movement exploitation strategies to balance the exploration capabilities of global search and the exploitation capabilities of local search.Simultaneously,the acceleration of convergence toward the global optimum and a reduction in computational complexity are facilitated by means of the global optimum bootstrap strategy and a linear population reduction strategy.Through experimental evaluation with multiple algorithms and diverse performance metrics,the proposed algorithm confirms reliable accuracy on three datasets:KDD CUP99,NSL_KDD,and UNSW_NB15,achieving accuracy of 98.57%,98.81%,and 98.32%,respectively.These results affirm its potential as an effective solution for practical clustering applications.
基金Support by National Natural Science Foundation Grant(Correlation Analysis Among Herb's Properties affecting the Multiple Parameters of the Blood and the Morphology of the Organism Structure,No.81473366)
文摘OBJECTIVE:To study the morphological basis of the role of Siqi(cold as winter,cool as autumn,warm as spring,hot as summer),Wuwei(five flavors:sweet,pungent,salty,sour,and bitter),and Guijing(meridian tropism) through the use of information integration.METHODS:A14C-2-deoxy-glucose autoradiography method was adopted to determine the overall impact of treatment with 39 herbs on functions of various tissues and organs.Data was measured at 4hs after a single dose and following the last treatment of repeated doses for a week.Least-squares estimation was used and fitted for each herb regression effect of organs and tissues after singleand repeated treatment.The slope of the regression line represented the cumulative trend of the effect of the herbs(β),and the standard deviation of the slope(Sβ) was compared with those of the untreated animals(t 'test).All significantly cumulative effect trends were applied with an artificial neural network,which integrated the relationship among Siqi,Wuwei,and Guijing with tissues and organs.RESULTS:There is a certain relationship among the Siqi,Wuwei,Guijing and the anatomy of organs and tissues,but the different scores indicate that influence of Siqi,Wuwei,Guijing to anatomy of organs and tissues was a nonlinear state.CONCLUSION:Results demonstrated that the effects of Siqi,Wuwei,and Guijing have a morphological basis,and each concept was associated with multiple anatomical structures.
基金Supported by Business Construction Project of China Meteorological Administration in 2011Key Project of Key Meteorological Technology Integration and Application of China Meteorological Administration(CMAGJ2013Z01)
文摘This paper has designed a unified storage model of Automatic Weather Station Quality Information Inquiries and Feedback( AWSQIF),and has developed an AWSQIF automatic management system,which implements the storage of AWSQIF automatically in national quality information two-way feedback process. Based on the unified storage,AWSQIF data have important application in real-time data management,data quality assessment and quality control effect analysis. The results show that the data quality has been improved significantly,and the data assessment has achieved online operating. In addition,AWSQIF can assist the improvement of the quality control effect.
基金the National High Technology Research and Development Program of China (863 Program) (No. 2008AA121600)the National BasicResearch Program of China (973 Program)(No. 2010CB731801)the National Natural Science Foundation of China (No. 40871212)
文摘Aiming at the integrative management and comprehensive applications of large-scale 3D geospatial information covering the full 3D space of a city, this paper briefly introduces the design and implementation of a full 3D GIS platform： GeoScope, which provides a professional solution for the massive full three-dimensional geospatial data integration, management, analysis, visualization, and applications. GeoScope is characterized by： （1） extendible software architecture based on the hierarchical message bus, facilitates multimodal integrative applications of 2D GIS and 3D GIS; （2） unified 3D city models, support multiscale se- mantic representation of outdoor ＆ indoor and aboveground ＆ underground 3D objects; （3） high-efficient 3D geospatial database engine, supports integrated management of massive 3D geospatial data for real-time applications; and （4） high-performance visualization engine exploiting the massively parallel computation architecture of modem GPUs, supports real-time realistic rendering of large-scale complicated 3D geospatial environments. The successful pilot application of GeoScope is also illustrated with the 3D city models of 8494 knl2 of the whole Wuhan City, the largest city in middle China.
基金the Hi-Tech and Research Development Program of China(2005AA133070)the Electronics Information Industry Development Fund Tender Project(XDJ2-0514-27)
文摘Analyzed the main problems which were found in current conditions and prob- lems of informationization in coal enterprises.It clarified how to achieve informationization in coal mine and put forward a general configuration of informationization construction in which informationization in coal enterprises was divided into two parts: informationization of safety production and informationization of management.Planned a platform of inte- grated management of informationization in coal enterprises.Ultimately,it has brought forward that an overall integrated digital mine is the way to achieve the goal of informa- tionization in coal enterprises,which can promote the application of automation,digitaliza- tion,networking,informaitionization to intellectualization.At the same time,the competi- tiveness of enterprises can be improved entirely,and new type of coal industry can be supported by information technology.
基金funded by the Project of"973"Program of China under contract No.2006 CB701305the National Natural Science Foundation of China under contract No.40571129.
文摘Marine geographic information system （MGIS） has great ability to deal with the spatio-temporal problems and has potential superiority when it is applied to oceanography. Using the feature extraction of oceanic phenomena as a case study, the functions of the MGIS are analyzed, and thus the position of MGIS in the oceanography is defined. Comparing the requirement of MGIS with that of the traditional GIS which has been developed in the terrestrial applications in the past four decades, the frame for the functions of MGIS is constructed. According to the established MGIS, some key technologies are discussed in detail with emphasis on the specialities which can distinguish the MGIS from the traditional GIS.
文摘During the 12^(th)Five-Year Plan period,the environmental protection information work made remarkable achievements,and the construction of nationwide air quality monitoring network was completed; the trinitarian monitoring system was established,and environmental emergency command ability enhanced significantly. However,the issues of " information island" and " application silo" are still outstanding,and the application and promotion of new technologies make slow progress; the capacity of information system to support environmental protection business is insufficient,and there is no capacity for decision analysis. In the 13^(th)Five-Year Plan period,it is needed to strengthen the application and promotion of new technologies such as cloud computing,big data,Internet of Things,and spatial information,mainly construct and improve " Cloud Computing Based Environmental Protection( CCBEP) ", " big data platform of ecological environment",and then realize the core objective of improving environmental quality.
