1 When the song Heart on My Sleeve first dropped,fans were on cloud nine,fully convinced that their favorite musicians had just dropped a surprise cooperation.But as the song racked up(累计)millions of streams online,...1 When the song Heart on My Sleeve first dropped,fans were on cloud nine,fully convinced that their favorite musicians had just dropped a surprise cooperation.But as the song racked up(累计)millions of streams online,the truth surfaced-the track was created by an Internet user who used artificial intelligence(AI)to perfectly mimic(模仿)human artists'voices.展开更多
To address the high-quality forged videos,traditional approaches typically have low recognition accuracy and tend to be easily misclassified.This paper tries to address the challenge of detecting high-quality deepfake...To address the high-quality forged videos,traditional approaches typically have low recognition accuracy and tend to be easily misclassified.This paper tries to address the challenge of detecting high-quality deepfake videos by promoting the accuracy of Artificial Intelligence Generated Content(AIGC)video authenticity detection with a multimodal information fusion approach.First,a high-quality multimodal video dataset is collected and normalized,including resolution correction and frame rate unification.Next,feature extraction techniques are employed to draw out features from visual,audio,and text modalities.Subsequently,these features are fused into a multilayer perceptron and attention mechanisms-based multimodal feature matrix.Finally,the matrix is fed into a multimodal information fusion layer in order to construct and train a deep learning model.Experimental findings show that the multimodal fusion model achieves an accuracy of 93.8%for the detection of video authenticity,showing significant improvement against other unimodal models,as well as affirming better performance and resistance of the model to AIGC video authenticity detection.展开更多
Evolutionary algorithms have been extensively utilized in practical applications.However,manually designed population updating formulas are inherently prone to the subjective influence of the designer.Genetic programm...Evolutionary algorithms have been extensively utilized in practical applications.However,manually designed population updating formulas are inherently prone to the subjective influence of the designer.Genetic programming(GP),characterized by its tree-based solution structure,is a widely adopted technique for optimizing the structure of mathematical models tailored to real-world problems.This paper introduces a GP-based framework(GPEAs)for the autonomous generation of update formulas,aiming to reduce human intervention.Partial modifications to tree-based GP have been instigated,encompassing adjustments to its initialization process and fundamental update operations such as crossover and mutation within the algorithm.By designing suitable function sets and terminal sets tailored to the selected evolutionary algorithm,and ultimately derive an improved update formula.The Cat Swarm Optimization Algorithm(CSO)is chosen as a case study,and the GP-EAs is employed to regenerate the speed update formulas of the CSO.To validate the feasibility of the GP-EAs,the comprehensive performance of the enhanced algorithm(GP-CSO)was evaluated on the CEC2017 benchmark suite.Furthermore,GP-CSO is applied to deduce suitable embedding factors,thereby improving the robustness of the digital watermarking process.The experimental results indicate that the update formulas generated through training with GP-EAs possess excellent performance scalability and practical application proficiency.展开更多
In this study, we use the Bohai Sea area as an example to investigate the characteristics of secondary microseisms and their impact on seismic noise based on the temporal frequency spectral analysis of observation dat...In this study, we use the Bohai Sea area as an example to investigate the characteristics of secondary microseisms and their impact on seismic noise based on the temporal frequency spectral analysis of observation data from 33 broadband seismic stations during strong gust periods, and new perspectives are proposed on the generation mechanisms of secondary microseisms. The results show that short-period double- frequency (SPDF) and long-period double-frequency (LPDF) microseisms exhibit significant alternating trends of strengthening and weakening in the northwest area of the Bohai Sea. SPDF microseisms are generated by irregular wind waves during strong off shore wind periods, with a broad frequency band distributed in the range of 0.2-1 Hz;LPDF microseisms are generated by regular swells during periods of sea wind weakening, with a narrow frequency band concentrated between 0.15 and 0.3 Hz. In terms of temporal dimensions, as the sea wind weakens, the energy of SPDF microseisms weakens, and the dominant frequencies increase, whereas the energy of LPDF microseisms strengthens and the dominant frequencies decrease, which is consistent with the process of the decay of wind waves and the growth of swells. In terms of spatial dimensions, as the microseisms propagate inland areas, the advantageous frequency band and energy of SPDF microseisms are reduced and significantly attenuated, respectively, whereas LPDF microseisms show no significant changes. And during the propagation process in high-elevation areas, LPDF microseisms exhibit a certain site amplifi cation eff ect when the energy is strong. The results provide important supplements to the basic theory of secondary microseisms, preliminarily reveal the relationship between the atmosphere, ocean, and seismic noise, and provide important theoretical references for conducting geological and oceanographic research based on the characteristics of secondary microseisms.展开更多
Domain Generation Algorithms(DGAs)continue to pose a significant threat inmodernmalware infrastructures by enabling resilient and evasive communication with Command and Control(C&C)servers.Traditional detection me...Domain Generation Algorithms(DGAs)continue to pose a significant threat inmodernmalware infrastructures by enabling resilient and evasive communication with Command and Control(C&C)servers.Traditional detection methods-rooted in statistical heuristics,feature engineering,and shallow machine learning-struggle to adapt to the increasing sophistication,linguistic mimicry,and adversarial variability of DGA variants.The emergence of Large Language Models(LLMs)marks a transformative shift in this landscape.Leveraging deep contextual understanding,semantic generalization,and few-shot learning capabilities,LLMs such as BERT,GPT,and T5 have shown promising results in detecting both character-based and dictionary-based DGAs,including previously unseen(zeroday)variants.This paper provides a comprehensive and critical review of LLM-driven DGA detection,introducing a structured taxonomy of LLM architectures,evaluating the linguistic and behavioral properties of benchmark datasets,and comparing recent detection frameworks across accuracy,latency,robustness,and multilingual performance.We also highlight key limitations,including challenges in adversarial resilience,model interpretability,deployment scalability,and privacy risks.To address these gaps,we present a forward-looking research roadmap encompassing adversarial training,model compression,cross-lingual benchmarking,and real-time integration with SIEM/SOAR platforms.This survey aims to serve as a foundational resource for advancing the development of scalable,explainable,and operationally viable LLM-based DGA detection systems.展开更多
In the era of big data,data-driven technologies are increasingly leveraged by industry to facilitate autonomous learning and intelligent decision-making.However,the challenge of“small samples in big data”emerges whe...In the era of big data,data-driven technologies are increasingly leveraged by industry to facilitate autonomous learning and intelligent decision-making.However,the challenge of“small samples in big data”emerges when datasets lack the comprehensive information necessary for addressing complex scenarios,which hampers adaptability.Thus,enhancing data completeness is essential.Knowledge-guided virtual sample generation transforms domain knowledge into extensive virtual datasets,thereby reducing dependence on limited real samples and enabling zero-sample fault diagnosis.This study used building air conditioning systems as a case study.We innovatively used the large language model(LLM)to acquire domain knowledge for sample generation,significantly lowering knowledge acquisition costs and establishing a generalized framework for knowledge acquisition in engineering applications.This acquired knowledge guided the design of diffusion boundaries in mega-trend diffusion(MTD),while the Monte Carlo method was used to sample within the diffusion function to create information-rich virtual samples.Additionally,a noise-adding technique was introduced to enhance the information entropy of these samples,thereby improving the robustness of neural networks trained with them.Experimental results showed that training the diagnostic model exclusively with virtual samples achieved an accuracy of 72.80%,significantly surpassing traditional small-sample supervised learning in terms of generalization.This underscores the quality and completeness of the generated virtual samples.展开更多
In this paper .an important property layer-preserving of LF group and an concept LF λ-subgroup generated by LF subset will be introduced. On the height of LF group theory ,we recognize the concept generated subgroup ...In this paper .an important property layer-preserving of LF group and an concept LF λ-subgroup generated by LF subset will be introduced. On the height of LF group theory ,we recognize the concept generated subgroup more clearly.展开更多
Currently,some photorealistic computer graphics are very similar to photographic images.Photorealistic computer generated graphics can be forged as photographic images,causing serious security problems.The aim of this...Currently,some photorealistic computer graphics are very similar to photographic images.Photorealistic computer generated graphics can be forged as photographic images,causing serious security problems.The aim of this work is to use a deep neural network to detect photographic images(PI)versus computer generated graphics(CG).In existing approaches,image feature classification is computationally intensive and fails to achieve realtime analysis.This paper presents an effective approach to automatically identify PI and CG based on deep convolutional neural networks(DCNNs).Compared with some existing methods,the proposed method achieves real-time forensic tasks by deepening the network structure.Experimental results show that this approach can effectively identify PI and CG with average detection accuracy of 98%.展开更多
This paper investigates the control role of the relative phase between the probe and driving fields on the gain, dispersion and populations in an open A system with spontaneously generated coherence （SGC）. It shows ...This paper investigates the control role of the relative phase between the probe and driving fields on the gain, dispersion and populations in an open A system with spontaneously generated coherence （SGC）. It shows that by adjusting the value of the relative phase, a change from lasing with inversion to lasing without inversion can be realized; the values and frequency spectrum regions of the inversionless gain and dispersion can be obviously varied; high refractive index with zero absorption and electromagnetically induced transparency can be achieved. It is also found that when the driving field is resonant, the shapes of the dispersion and the gain curves versus the probe detuning are very similar if the relative phase of the dispersion lags π/2 than that of the gain, however for the off-resonant driving field the similarity will disappear; the gain, dispersion and populations are periodical functions of the relative phase, the modulation period is always 2π; the contribution of SGC to the inversionless gain and dispersion is much larger than that of the dynamically induced coherence.展开更多
We have studied the effect of the spontaneously generated coherence （SGC） on gain of lasing without inversion （LWI） in a closed three-level A-type atomic system with Doppler broadening. It is shown that, regardles...We have studied the effect of the spontaneously generated coherence （SGC） on gain of lasing without inversion （LWI） in a closed three-level A-type atomic system with Doppler broadening. It is shown that, regardless of the driving and probe fields being co- or counter-propagating, at a suitable value of the Doppler width, we can obtain a much larger LWI gain with SGC than that without SGC; and the region of the LWI gain spectrum with SGC is obviously larger than that without SGC. When the Doppler width takes a constant value, the gain does not monotonically decrease or increase with increasing strength of SGC, the largest LWI gain can be obtained by adjusting strength of SGC. Generally speaking, the co-propagating probe and driving fields is favourable to obtain a larger LWI gain.展开更多
An OpenFOAM based turbulence combustion solver with flamelet generated manifolds (FGMs) is presented in this paper. A series of flamelets, representative for turbulent flames, are calculated first by a one-dimensional...An OpenFOAM based turbulence combustion solver with flamelet generated manifolds (FGMs) is presented in this paper. A series of flamelets, representative for turbulent flames, are calculated first by a one-dimensional (1D) detailed chemistry solver with the consideration of both transport and stretch/curvature contributions. The flame structure is then parameterized as a function of multiple reaction control variables. A manifold, which collects the 1D flame properties, is built from the 1D flame solutions. The control variables of the mixture fraction and the progress variable are solved from the corresponding transport equations. During the calculation, the scalar variables, e.g., temperature and species concentration, are retrieved from the manifolds by interpolation. A transport equation for NO is solved to improve its prediction accuracy. To verify the ability to deal with the enthalpy loss effect, the temperature retrieved directly from the manifolds is compared with the temperature solved from a transport equation of absolute enthalpy. The resulting FGM-computational fluid dynamics (CFD) coupled code has three significant features, i.e., accurate NO prediction, the ability to treat the heat loss effect and the adoption at the turbulence level, and high quality prediction within practical industrial configurations. The proposed method is validated against the Sandia flame D, and good agreement with the experimental data is obtained.展开更多
We present a typhoon-generated noise model with which the noise intensity during typhoons can be estimated accurately. The model is verified through experimental study, and the simulation results agree reasonably with...We present a typhoon-generated noise model with which the noise intensity during typhoons can be estimated accurately. The model is verified through experimental study, and the simulation results agree reasonably with the experimental data. The measured noise intensity is approximately proportional to the cube of the local wind speed.展开更多
The low frequency load of an underwater explosion bubble and the generated waves can cause significant rigid motion of a ship that threaten its stability.In order to study the fluid-structure interaction qualitatively...The low frequency load of an underwater explosion bubble and the generated waves can cause significant rigid motion of a ship that threaten its stability.In order to study the fluid-structure interaction qualitatively,a two-dimensional underwater explosion bubble dynamics model,based on the potential flow theory,is established with a double-vortex model for the doubly connected bubble dynamics simulation,and the bubble shows similar dynamics to that in 3-dimensional domain.A fully nonlinear fluid-structure interaction model is established considering the rigid motion of the floating body using the mode-decomposition method.Convergence test of the model is implemented by simulating the free rolling motion of a floating body in still water.Through the simulation of the interaction of the underwater explosion bubble,the generated waves and the floating body based on the presented model,the influences of the buoyancy parameter and the distance parameter are discussed.It is found that the impact loads on floating body caused by underwater explosion bubble near the free surface can be divided into 3 components:bubble pulsation,jet impact,and slamming load of the generated waves,and the intensity of each component changes nonlinearly with the buoyance parameter.The bubble pulsation load decays with the increase in the horizontal distance.However,the impact load from the generated waves is not monotonous to distance.It increases with the distance within a particular distance threshold,but decays thereafter.展开更多
The use of non-power frequency components for protection measurements in an area that has received relatively little attention. Although work on the use of switched stack tuners has been reported for fault detection, ...The use of non-power frequency components for protection measurements in an area that has received relatively little attention. Although work on the use of switched stack tuners has been reported for fault detection, the sensitivity of switched arrangements is limited by scnsitivity of directional detectors for switching purposes. This paper presents an alternative non-switching scheme which is relatively simple.A simplified typical 400kV system with the line trap, stack tuner, coupling capacitor and bus capacitance to earth is considered. Fault studies by using the EMTP have been carried out, frequency-dependent pararnctcrs are used for the line representation. Arcing faults have also bccn investigated and the results show a great promise of the scheme.展开更多
Tourism-led economic growth and tourism-driven urbanization have attracted increasing attention by provinces and regions in China with abundant tourism resources.Due to low data availability,the current tourism litera...Tourism-led economic growth and tourism-driven urbanization have attracted increasing attention by provinces and regions in China with abundant tourism resources.Due to low data availability,the current tourism literature lacks empirical evidence of the tourism network in lessdeveloped mountainous regions where the development of transport infrastructure is more variable.This paper aims to provide such evidence using Guangxi Zhuang Autonomous Region in China as a case study.Using User Generated Content(UGC)data,this study constructs a tourism network in Guangxi.By integrating social network analysis with spatial interaction modelling,we compared the impact of two different transport infrastructures,highway and high-speed railway,on tourist flows,particularly in less-developed mountainous regions.It was found that the product of node centrality and flow could best describe the significant pushing and pulling forces on the flow of tourists.The tourism by high-speed railway was sensitive to the position of trip destination on the whole tourism network but self-drive tourism was more sensitive to travelling time.The increase of high-speed railway density is crucial to promote local tourism-led economic development,however,large-scale karst landforms in the study area present a significant obstacle to the construction of high-speed railways.展开更多
In this paper we study influences of Doppler broadening, spontaneously generated coherence, and other system parameters on propagation effect in a quasi lambda-type four-level atomic system. It is shown that when the ...In this paper we study influences of Doppler broadening, spontaneously generated coherence, and other system parameters on propagation effect in a quasi lambda-type four-level atomic system. It is shown that when the Doppler broadening is present, generally speaking, the values of gain and intensity of lasing without inversion （i.e. the probe field） in the co-propagating probe and driving fields case are much larger than those in the counter-propagating case; considerably larger gain and intensity of lasing without inversion than those without the Doppler broadening can be obtained by choosing appropriate values of the Doppler broadening width and spontaneously generated coherence strength. The gain and intensity of lasing without inversion increase with the increase of spontaneously generated coherence strength; when spontaneously generated coherence is present, much larger gain and intensity of lasing without inversion than those in the case without spontaneously generated coherence can be obtained. Choosing suitable values of the probe detuning, Rabi frequencies of the driving and pump fields at the entrance of the medium also can remarkably enhance the gain and intensity of lasing without inversion.展开更多
The potential of carrying out oxidative desulfurization(ODS) using oxygen as an oxidant was explored in this work. n-Octane firstly reacted with oxygen to produce hydroperoxides in-situ, which were then used as oxidan...The potential of carrying out oxidative desulfurization(ODS) using oxygen as an oxidant was explored in this work. n-Octane firstly reacted with oxygen to produce hydroperoxides in-situ, which were then used as oxidants to oxidize the dibenzothiophene(DBT) in the absence of catalysts. The hydroperoxides generated in-situ were effective in oxidizing DBT to its corresponding dibenzothiophene sulfone(DBTO_2) which was characterized by FT-IR and ~1H-NMR. The removal rate of DBT could reached 98.4% under conditions covering a temperaure of 140℃, a rection duration of 4 h, and an oxygen partial pressure of 0.4 MPa. The influences of different hydrocarbon components in diesel on DBT removal were investigated. The results showed that cyclohexane and n-dodecane had no effect on the removal of DBT, but xylene had a slight negative effect on DBT removal. A possible oxidation mechanism was proposed and the concentration of hydroperoxides in both O_2-oxidized octane and model diesel were detected.展开更多
It is challenging to measure the electron density of the unsteady plasma formed by charged particles generated from explosions in the air,because it is transient and on a microsecond time scale.In this study,the time-...It is challenging to measure the electron density of the unsteady plasma formed by charged particles generated from explosions in the air,because it is transient and on a microsecond time scale.In this study,the time-varying electron density of the plasma generated from a small cylindrical cyclotrimethylenetrinitramine(RDX)explosion in air was measured,based on the principle of microwave Rayleigh scattering.It was found that the evolution of the electron density is related to the diffusion of the detonation products.The application of the Rayleigh microwave scattering principle is an attempt to estimate the electron density in explosively generated plasma.Using the equivalent radius and length of the detonation products in the bright areas of images taken by a high-speed framing camera,the electron density was determined to be of the order of 10^(20)m^(−3).The delay time between the initiation time and the start of variation in the electron-density curve was 2.77–6.93μs.In the time-varying Rayleigh microwave scattering signal curve of the explosively generated plasma,the electron density had two fluctuation processes.The durations of the first stage and the second stage were 11.32μs and 19.20μs,respectively.Both fluctuation processes increased rapidly to a peak value and then rapidly attenuated with time.This revealed the movement characteristics of the charged particles during the explosion.展开更多
Brain plasticity is heterogeneous in mammals：Brain regeneration and repair are the dream of every neurobiologist as well as every common citizen in the world who knows that most neurological diseases,dementia and oth...Brain plasticity is heterogeneous in mammals：Brain regeneration and repair are the dream of every neurobiologist as well as every common citizen in the world who knows that most neurological diseases,dementia and other age-related problems affecting the central nervous system（CNS）do represent a heavy health and social burden.Efficacious re-generative processes are not＂ a natural property of the mammalian CNS, rather, due to evolutionary constraints they seem substantially reduced （if compared to those occurring in non-mammalian vertebrates） and hardly inducible by therapeutic approaches （reviewed in Martino et al., 2011）.展开更多
With the development of computer graphics,realistic computer graphics(CG)have become more and more common in our field of vision.This rendered image is invisible to the naked eye.How to effectively identify CG and nat...With the development of computer graphics,realistic computer graphics(CG)have become more and more common in our field of vision.This rendered image is invisible to the naked eye.How to effectively identify CG and natural images(NI)has been become a new issue in the field of digital forensics.In recent years,a series of deep learning network frameworks have shown great advantages in the field of images,which provides a good choice for us to solve this problem.This paper aims to track the latest developments and applications of deep learning in the field of CG and NI forensics in a timely manner.Firstly,it introduces the background of deep learning and the knowledge of convolutional neural networks.The purpose is to understand the basic model structure of deep learning applications in the image field,and then outlines the mainstream framework;secondly,it briefly introduces the application of deep learning in CG and NI forensics,and finally points out the problems of deep learning in this field and the prospects for the future.展开更多
文摘1 When the song Heart on My Sleeve first dropped,fans were on cloud nine,fully convinced that their favorite musicians had just dropped a surprise cooperation.But as the song racked up(累计)millions of streams online,the truth surfaced-the track was created by an Internet user who used artificial intelligence(AI)to perfectly mimic(模仿)human artists'voices.
文摘To address the high-quality forged videos,traditional approaches typically have low recognition accuracy and tend to be easily misclassified.This paper tries to address the challenge of detecting high-quality deepfake videos by promoting the accuracy of Artificial Intelligence Generated Content(AIGC)video authenticity detection with a multimodal information fusion approach.First,a high-quality multimodal video dataset is collected and normalized,including resolution correction and frame rate unification.Next,feature extraction techniques are employed to draw out features from visual,audio,and text modalities.Subsequently,these features are fused into a multilayer perceptron and attention mechanisms-based multimodal feature matrix.Finally,the matrix is fed into a multimodal information fusion layer in order to construct and train a deep learning model.Experimental findings show that the multimodal fusion model achieves an accuracy of 93.8%for the detection of video authenticity,showing significant improvement against other unimodal models,as well as affirming better performance and resistance of the model to AIGC video authenticity detection.
文摘Evolutionary algorithms have been extensively utilized in practical applications.However,manually designed population updating formulas are inherently prone to the subjective influence of the designer.Genetic programming(GP),characterized by its tree-based solution structure,is a widely adopted technique for optimizing the structure of mathematical models tailored to real-world problems.This paper introduces a GP-based framework(GPEAs)for the autonomous generation of update formulas,aiming to reduce human intervention.Partial modifications to tree-based GP have been instigated,encompassing adjustments to its initialization process and fundamental update operations such as crossover and mutation within the algorithm.By designing suitable function sets and terminal sets tailored to the selected evolutionary algorithm,and ultimately derive an improved update formula.The Cat Swarm Optimization Algorithm(CSO)is chosen as a case study,and the GP-EAs is employed to regenerate the speed update formulas of the CSO.To validate the feasibility of the GP-EAs,the comprehensive performance of the enhanced algorithm(GP-CSO)was evaluated on the CEC2017 benchmark suite.Furthermore,GP-CSO is applied to deduce suitable embedding factors,thereby improving the robustness of the digital watermarking process.The experimental results indicate that the update formulas generated through training with GP-EAs possess excellent performance scalability and practical application proficiency.
基金supported by Earthquake Science and Technology Spark Program of China Earthquake Administration (No. XH20006Y)Local Standards Formulation and Revision Program of Hebei Province (No. FW202154)+1 种基金Earthquake Science and Technology Spark Program of Hebei Earthquake Agency (No. DZ2024112100002)2023 Seismological Data Sharing Project of China Earthquake Networks Center (Dataset Project)。
文摘In this study, we use the Bohai Sea area as an example to investigate the characteristics of secondary microseisms and their impact on seismic noise based on the temporal frequency spectral analysis of observation data from 33 broadband seismic stations during strong gust periods, and new perspectives are proposed on the generation mechanisms of secondary microseisms. The results show that short-period double- frequency (SPDF) and long-period double-frequency (LPDF) microseisms exhibit significant alternating trends of strengthening and weakening in the northwest area of the Bohai Sea. SPDF microseisms are generated by irregular wind waves during strong off shore wind periods, with a broad frequency band distributed in the range of 0.2-1 Hz;LPDF microseisms are generated by regular swells during periods of sea wind weakening, with a narrow frequency band concentrated between 0.15 and 0.3 Hz. In terms of temporal dimensions, as the sea wind weakens, the energy of SPDF microseisms weakens, and the dominant frequencies increase, whereas the energy of LPDF microseisms strengthens and the dominant frequencies decrease, which is consistent with the process of the decay of wind waves and the growth of swells. In terms of spatial dimensions, as the microseisms propagate inland areas, the advantageous frequency band and energy of SPDF microseisms are reduced and significantly attenuated, respectively, whereas LPDF microseisms show no significant changes. And during the propagation process in high-elevation areas, LPDF microseisms exhibit a certain site amplifi cation eff ect when the energy is strong. The results provide important supplements to the basic theory of secondary microseisms, preliminarily reveal the relationship between the atmosphere, ocean, and seismic noise, and provide important theoretical references for conducting geological and oceanographic research based on the characteristics of secondary microseisms.
基金the Deanship of Scientific Research at King Khalid University for funding this work through large group under grant number(GRP.2/663/46).
文摘Domain Generation Algorithms(DGAs)continue to pose a significant threat inmodernmalware infrastructures by enabling resilient and evasive communication with Command and Control(C&C)servers.Traditional detection methods-rooted in statistical heuristics,feature engineering,and shallow machine learning-struggle to adapt to the increasing sophistication,linguistic mimicry,and adversarial variability of DGA variants.The emergence of Large Language Models(LLMs)marks a transformative shift in this landscape.Leveraging deep contextual understanding,semantic generalization,and few-shot learning capabilities,LLMs such as BERT,GPT,and T5 have shown promising results in detecting both character-based and dictionary-based DGAs,including previously unseen(zeroday)variants.This paper provides a comprehensive and critical review of LLM-driven DGA detection,introducing a structured taxonomy of LLM architectures,evaluating the linguistic and behavioral properties of benchmark datasets,and comparing recent detection frameworks across accuracy,latency,robustness,and multilingual performance.We also highlight key limitations,including challenges in adversarial resilience,model interpretability,deployment scalability,and privacy risks.To address these gaps,we present a forward-looking research roadmap encompassing adversarial training,model compression,cross-lingual benchmarking,and real-time integration with SIEM/SOAR platforms.This survey aims to serve as a foundational resource for advancing the development of scalable,explainable,and operationally viable LLM-based DGA detection systems.
基金supported by the National Natural Science Foundation of China(No.62306281)the Natural Science Foundation of Zhejiang Province(Nos.LQ23E060006 and LTGG24E050005)the Key Research Plan of Jiaxing City(No.2024BZ20016).
文摘In the era of big data,data-driven technologies are increasingly leveraged by industry to facilitate autonomous learning and intelligent decision-making.However,the challenge of“small samples in big data”emerges when datasets lack the comprehensive information necessary for addressing complex scenarios,which hampers adaptability.Thus,enhancing data completeness is essential.Knowledge-guided virtual sample generation transforms domain knowledge into extensive virtual datasets,thereby reducing dependence on limited real samples and enabling zero-sample fault diagnosis.This study used building air conditioning systems as a case study.We innovatively used the large language model(LLM)to acquire domain knowledge for sample generation,significantly lowering knowledge acquisition costs and establishing a generalized framework for knowledge acquisition in engineering applications.This acquired knowledge guided the design of diffusion boundaries in mega-trend diffusion(MTD),while the Monte Carlo method was used to sample within the diffusion function to create information-rich virtual samples.Additionally,a noise-adding technique was introduced to enhance the information entropy of these samples,thereby improving the robustness of neural networks trained with them.Experimental results showed that training the diagnostic model exclusively with virtual samples achieved an accuracy of 72.80%,significantly surpassing traditional small-sample supervised learning in terms of generalization.This underscores the quality and completeness of the generated virtual samples.
文摘In this paper .an important property layer-preserving of LF group and an concept LF λ-subgroup generated by LF subset will be introduced. On the height of LF group theory ,we recognize the concept generated subgroup more clearly.
基金This work is supported,in part,by the National Natural Science Foundation of China under grant numbers U1536206,U1405254,61772283,61602253,61672294,61502242In part,by the Jiangsu Basic Research Programs-Natural Science Foundation under grant numbers BK20150925 and BK20151530+1 种基金In part,by the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)fundIn part,by the Collaborative Innovation Center of Atmospheric Environment and Equipment Technology(CICAEET)fund,China.
文摘Currently,some photorealistic computer graphics are very similar to photographic images.Photorealistic computer generated graphics can be forged as photographic images,causing serious security problems.The aim of this work is to use a deep neural network to detect photographic images(PI)versus computer generated graphics(CG).In existing approaches,image feature classification is computationally intensive and fails to achieve realtime analysis.This paper presents an effective approach to automatically identify PI and CG based on deep convolutional neural networks(DCNNs).Compared with some existing methods,the proposed method achieves real-time forensic tasks by deepening the network structure.Experimental results show that this approach can effectively identify PI and CG with average detection accuracy of 98%.
文摘This paper investigates the control role of the relative phase between the probe and driving fields on the gain, dispersion and populations in an open A system with spontaneously generated coherence （SGC）. It shows that by adjusting the value of the relative phase, a change from lasing with inversion to lasing without inversion can be realized; the values and frequency spectrum regions of the inversionless gain and dispersion can be obviously varied; high refractive index with zero absorption and electromagnetically induced transparency can be achieved. It is also found that when the driving field is resonant, the shapes of the dispersion and the gain curves versus the probe detuning are very similar if the relative phase of the dispersion lags π/2 than that of the gain, however for the off-resonant driving field the similarity will disappear; the gain, dispersion and populations are periodical functions of the relative phase, the modulation period is always 2π; the contribution of SGC to the inversionless gain and dispersion is much larger than that of the dynamically induced coherence.
基金Project supported by the National Natural Science Foundation of China （Grant No 10675076）, the Natural Science Foundation of Shandong Province, China （Grant No Y2006A21） and the State Key Laboratory of High Field Laser Physics, Shanghai Institute of 0ptics and Fine Mechanics, Chinese Academy of Sciences.
文摘We have studied the effect of the spontaneously generated coherence （SGC） on gain of lasing without inversion （LWI） in a closed three-level A-type atomic system with Doppler broadening. It is shown that, regardless of the driving and probe fields being co- or counter-propagating, at a suitable value of the Doppler width, we can obtain a much larger LWI gain with SGC than that without SGC; and the region of the LWI gain spectrum with SGC is obviously larger than that without SGC. When the Doppler width takes a constant value, the gain does not monotonically decrease or increase with increasing strength of SGC, the largest LWI gain can be obtained by adjusting strength of SGC. Generally speaking, the co-propagating probe and driving fields is favourable to obtain a larger LWI gain.
基金Project supported by the Strategic Priority Research Program of the Chinese Academy of Sciences(No.XDA 21060102)Guangdong Provincial Key Laboratory of New and Renewable Energy Research and Development of China(No.y809jh1001)
文摘An OpenFOAM based turbulence combustion solver with flamelet generated manifolds (FGMs) is presented in this paper. A series of flamelets, representative for turbulent flames, are calculated first by a one-dimensional (1D) detailed chemistry solver with the consideration of both transport and stretch/curvature contributions. The flame structure is then parameterized as a function of multiple reaction control variables. A manifold, which collects the 1D flame properties, is built from the 1D flame solutions. The control variables of the mixture fraction and the progress variable are solved from the corresponding transport equations. During the calculation, the scalar variables, e.g., temperature and species concentration, are retrieved from the manifolds by interpolation. A transport equation for NO is solved to improve its prediction accuracy. To verify the ability to deal with the enthalpy loss effect, the temperature retrieved directly from the manifolds is compared with the temperature solved from a transport equation of absolute enthalpy. The resulting FGM-computational fluid dynamics (CFD) coupled code has three significant features, i.e., accurate NO prediction, the ability to treat the heat loss effect and the adoption at the turbulence level, and high quality prediction within practical industrial configurations. The proposed method is validated against the Sandia flame D, and good agreement with the experimental data is obtained.
基金Supported by the National Natural Science Foundation of China under Grant No 11125420
文摘We present a typhoon-generated noise model with which the noise intensity during typhoons can be estimated accurately. The model is verified through experimental study, and the simulation results agree reasonably with the experimental data. The measured noise intensity is approximately proportional to the cube of the local wind speed.
基金This work was supported by the National Natural ScienceFoundation of China (Grant No. 51879050, 51609044), the Defense IndustrialTechnology Development Program of China (Grant No. JCKY2017604C002), NaturalScience Foundation of Heilongjiang Province of China (No. E2017021) and ShenzhenSpecial Fund for Future Industries (Grant No. JCYJ20160331163751413).
文摘The low frequency load of an underwater explosion bubble and the generated waves can cause significant rigid motion of a ship that threaten its stability.In order to study the fluid-structure interaction qualitatively,a two-dimensional underwater explosion bubble dynamics model,based on the potential flow theory,is established with a double-vortex model for the doubly connected bubble dynamics simulation,and the bubble shows similar dynamics to that in 3-dimensional domain.A fully nonlinear fluid-structure interaction model is established considering the rigid motion of the floating body using the mode-decomposition method.Convergence test of the model is implemented by simulating the free rolling motion of a floating body in still water.Through the simulation of the interaction of the underwater explosion bubble,the generated waves and the floating body based on the presented model,the influences of the buoyancy parameter and the distance parameter are discussed.It is found that the impact loads on floating body caused by underwater explosion bubble near the free surface can be divided into 3 components:bubble pulsation,jet impact,and slamming load of the generated waves,and the intensity of each component changes nonlinearly with the buoyance parameter.The bubble pulsation load decays with the increase in the horizontal distance.However,the impact load from the generated waves is not monotonous to distance.It increases with the distance within a particular distance threshold,but decays thereafter.
文摘The use of non-power frequency components for protection measurements in an area that has received relatively little attention. Although work on the use of switched stack tuners has been reported for fault detection, the sensitivity of switched arrangements is limited by scnsitivity of directional detectors for switching purposes. This paper presents an alternative non-switching scheme which is relatively simple.A simplified typical 400kV system with the line trap, stack tuner, coupling capacitor and bus capacitance to earth is considered. Fault studies by using the EMTP have been carried out, frequency-dependent pararnctcrs are used for the line representation. Arcing faults have also bccn investigated and the results show a great promise of the scheme.
基金funded by the Guangxi Natural Science Foundation(Grant No.2020GXNSFAA159065)the Opening Fund of Key Laboratory of Environment Change and Resources Use in Beibu Gulf under Ministry of Education(Nanning Normal University)+1 种基金Guangxi Key Laboratory of Earth Surface Processes and Intelligent Simulation(Nanning Normal University)(Grant No.GTEU-KLOP-K1701)the seventh batch of distinguished experts in Guangxi and National Natural Science Foundation of China(Grant No.41867071)。
文摘Tourism-led economic growth and tourism-driven urbanization have attracted increasing attention by provinces and regions in China with abundant tourism resources.Due to low data availability,the current tourism literature lacks empirical evidence of the tourism network in lessdeveloped mountainous regions where the development of transport infrastructure is more variable.This paper aims to provide such evidence using Guangxi Zhuang Autonomous Region in China as a case study.Using User Generated Content(UGC)data,this study constructs a tourism network in Guangxi.By integrating social network analysis with spatial interaction modelling,we compared the impact of two different transport infrastructures,highway and high-speed railway,on tourist flows,particularly in less-developed mountainous regions.It was found that the product of node centrality and flow could best describe the significant pushing and pulling forces on the flow of tourists.The tourism by high-speed railway was sensitive to the position of trip destination on the whole tourism network but self-drive tourism was more sensitive to travelling time.The increase of high-speed railway density is crucial to promote local tourism-led economic development,however,large-scale karst landforms in the study area present a significant obstacle to the construction of high-speed railways.
基金supported by the National Natural Science Foundation of China (Grant No.10875072)
文摘In this paper we study influences of Doppler broadening, spontaneously generated coherence, and other system parameters on propagation effect in a quasi lambda-type four-level atomic system. It is shown that when the Doppler broadening is present, generally speaking, the values of gain and intensity of lasing without inversion （i.e. the probe field） in the co-propagating probe and driving fields case are much larger than those in the counter-propagating case; considerably larger gain and intensity of lasing without inversion than those without the Doppler broadening can be obtained by choosing appropriate values of the Doppler broadening width and spontaneously generated coherence strength. The gain and intensity of lasing without inversion increase with the increase of spontaneously generated coherence strength; when spontaneously generated coherence is present, much larger gain and intensity of lasing without inversion than those in the case without spontaneously generated coherence can be obtained. Choosing suitable values of the probe detuning, Rabi frequencies of the driving and pump fields at the entrance of the medium also can remarkably enhance the gain and intensity of lasing without inversion.
基金the Undergraduate Innovation and Entrepreneurship Training Project (201710057009) for providing funding and support for this research
文摘The potential of carrying out oxidative desulfurization(ODS) using oxygen as an oxidant was explored in this work. n-Octane firstly reacted with oxygen to produce hydroperoxides in-situ, which were then used as oxidants to oxidize the dibenzothiophene(DBT) in the absence of catalysts. The hydroperoxides generated in-situ were effective in oxidizing DBT to its corresponding dibenzothiophene sulfone(DBTO_2) which was characterized by FT-IR and ~1H-NMR. The removal rate of DBT could reached 98.4% under conditions covering a temperaure of 140℃, a rection duration of 4 h, and an oxygen partial pressure of 0.4 MPa. The influences of different hydrocarbon components in diesel on DBT removal were investigated. The results showed that cyclohexane and n-dodecane had no effect on the removal of DBT, but xylene had a slight negative effect on DBT removal. A possible oxidation mechanism was proposed and the concentration of hydroperoxides in both O_2-oxidized octane and model diesel were detected.
基金supported by National Natural Science Foundation of China(Nos.11502118,11504173).
文摘It is challenging to measure the electron density of the unsteady plasma formed by charged particles generated from explosions in the air,because it is transient and on a microsecond time scale.In this study,the time-varying electron density of the plasma generated from a small cylindrical cyclotrimethylenetrinitramine(RDX)explosion in air was measured,based on the principle of microwave Rayleigh scattering.It was found that the evolution of the electron density is related to the diffusion of the detonation products.The application of the Rayleigh microwave scattering principle is an attempt to estimate the electron density in explosively generated plasma.Using the equivalent radius and length of the detonation products in the bright areas of images taken by a high-speed framing camera,the electron density was determined to be of the order of 10^(20)m^(−3).The delay time between the initiation time and the start of variation in the electron-density curve was 2.77–6.93μs.In the time-varying Rayleigh microwave scattering signal curve of the explosively generated plasma,the electron density had two fluctuation processes.The durations of the first stage and the second stage were 11.32μs and 19.20μs,respectively.Both fluctuation processes increased rapidly to a peak value and then rapidly attenuated with time.This revealed the movement characteristics of the charged particles during the explosion.
基金supported by MIUR-PRIN2015(grant 2015Y5W9YP)University of Turin(Ricerca locale 2016)
文摘Brain plasticity is heterogeneous in mammals：Brain regeneration and repair are the dream of every neurobiologist as well as every common citizen in the world who knows that most neurological diseases,dementia and other age-related problems affecting the central nervous system（CNS）do represent a heavy health and social burden.Efficacious re-generative processes are not＂ a natural property of the mammalian CNS, rather, due to evolutionary constraints they seem substantially reduced （if compared to those occurring in non-mammalian vertebrates） and hardly inducible by therapeutic approaches （reviewed in Martino et al., 2011）.
基金supported by National Natural Science Foundation of China(62072250).
文摘With the development of computer graphics,realistic computer graphics(CG)have become more and more common in our field of vision.This rendered image is invisible to the naked eye.How to effectively identify CG and natural images(NI)has been become a new issue in the field of digital forensics.In recent years,a series of deep learning network frameworks have shown great advantages in the field of images,which provides a good choice for us to solve this problem.This paper aims to track the latest developments and applications of deep learning in the field of CG and NI forensics in a timely manner.Firstly,it introduces the background of deep learning and the knowledge of convolutional neural networks.The purpose is to understand the basic model structure of deep learning applications in the image field,and then outlines the mainstream framework;secondly,it briefly introduces the application of deep learning in CG and NI forensics,and finally points out the problems of deep learning in this field and the prospects for the future.
