地球大气层外太阳光谱辐照度(extraterrestrial solar spectral irradiance,ESSI)数据是计算卫星传感器波段平均太阳辐照度(band mean solar irradiance,BMSI)的重要参数。为了探求利用何种来源的ESSI数据计算传感器BMSI更为准确,分别采...地球大气层外太阳光谱辐照度(extraterrestrial solar spectral irradiance,ESSI)数据是计算卫星传感器波段平均太阳辐照度(band mean solar irradiance,BMSI)的重要参数。为了探求利用何种来源的ESSI数据计算传感器BMSI更为准确,分别采用SBDART软件模拟的太阳光谱曲线数据、MODTRAN4.0 oldkur.dat文件数据、Thuillier太阳光谱曲线数据和WRC太阳光谱曲线数据计算了HJ-1A CCD1(B1—B4),CBERS-02 CCD(B1—B5),Landsat5TM(B1—B4)和ASTER(B1—B8)4种传感器的BMSI,并与传感器运营商公布的数据进行了比较。结果表明:利用SBDART和WRC太阳光谱曲线数据计算的结果误差较小;利用MODTRAN4.0 oldkur.dat数据计算的结果误差次之;利用Thuillier太阳光谱曲线的计算结果误差较大。展开更多
目的总结1例Rh系统抗-c胎儿新生儿溶血病(hemolytic disease of the fetus and newborn,HDFN)的实验室检测并进行文献复习,探究抗-c HDFN的特点。方法采用血清学方法检测患儿及其母亲的ABO血型、Rh分型、直接抗人球蛋白试验(DAT)、不规...目的总结1例Rh系统抗-c胎儿新生儿溶血病(hemolytic disease of the fetus and newborn,HDFN)的实验室检测并进行文献复习,探究抗-c HDFN的特点。方法采用血清学方法检测患儿及其母亲的ABO血型、Rh分型、直接抗人球蛋白试验(DAT)、不规则抗体及其效价,并对本实验室及国内外抗-c HDFN案例进行统计,分析比较抗-c与抗-D、抗-E引起HDFN的严重率。结果患儿血型为B、CcDee,DAT阳性,血清和放散液均检测出抗-c,血清抗体效价为4;母亲血型为AB、CCDee,DAT阴性,血清中检测出抗-c,效价为128。20例抗-c HDFN案例中DAT阳性有17例,输血或换血的有9例,占45%(9/20)。严重率为:抗-c 47.60%(10/21)、抗-D 47.60%(10/21)、抗-E 31.30%(5/16)。结论母亲妊娠和/或输血是产生抗-c等Rh系统同种抗体的主要原因。抗-c的预防管理应类同抗-D,对育龄女性输血要做好Rh血型5种抗原匹配输注以避免抗体产生,并重视孕妇产检Rh血型鉴定和抗体筛查鉴定,做到早发现、早干预、早治疗。展开更多
谱聚类因其在建模数据间成对相似关系方面的优越性而广泛应用于无监督学习领域.然而,传统谱聚类方法通常依赖干净、结构一致的数据分布,在现实应用中面临常见的噪声样本时,性能显著下降.针对该问题,文中提出融合CLIP(Contrastive Langua...谱聚类因其在建模数据间成对相似关系方面的优越性而广泛应用于无监督学习领域.然而,传统谱聚类方法通常依赖干净、结构一致的数据分布,在现实应用中面临常见的噪声样本时,性能显著下降.针对该问题,文中提出融合CLIP(Contrastive Language Image Pretraining)先验知识的谱聚类框架——基于知识重用的噪声环境谱聚类(Noise Spectral Clustering with Assistance of Knowledge Reuse,NSCR).该方法充分利用多模态神经网络在跨模态语义理解上的先验能力,构建基于知识重用的伪标签生成机制,通过多模型语义一致性判别机制与基于信息熵的不确定性建模机制识别高可信样本.同时引入归一化指数熵作为伪标签不确定性度量指标,从多模型输出中筛选语义一致、信息熵较低的样本,并生成伪标签,监督信号形式,引导聚类过程.此外,引入联合优化目标,扩展传统谱聚类方法,通过特征对齐与正则化平衡因子缓解伪标签监督与聚类目标之间的语义冲突.在多个公开数据集上的实验表明,NSCR在不同类型噪声干扰下的鲁棒性与泛化性良好.展开更多
Background:Although there is growing evidence of the use of artificial intelligence(AI)techniques in sports,ethical issues surrounding AI use are being discussed at a minimal level.Thus,this systematic scoping review ...Background:Although there is growing evidence of the use of artificial intelligence(AI)techniques in sports,ethical issues surrounding AI use are being discussed at a minimal level.Thus,this systematic scoping review aimed to summarize the current ethical implications associated with using AI in sports.Methods:In this study,a total of 9 databases-MEDLINE/PubMed,Embase,Cochrane Library,ProQuest,EBSCOhost,IEEE Xplore,Web of Science,Scopus,and Google Scholar--were searched.The review protocol was registered(https://osfio/42a8q)before extracting data.The search yielded 397 studies,and 25 studies met the inclusion and exclusion criteria.Results:The 25 studies were categorized into 4 primary ethical concerns:fairness and bias,transparency and explainability,privacy and data ethics,and accountability in AI's application in sports.These categorizations were derived based on the systematic review ofethical issues highlighted across the selected studies.Fifteen studies delved into fairness and bias,focusing on how AI can perpetuate existing inequalities in sports.Thirteen studies addressed the lack of transparency,emphasizing the challenges in interpretability and trust in AI-driven decisions.Privacy and data ethics emerged as significant in22 studies,highlighting risks related to the misuse of athletes’sensitive data.Finally,account-ability was examined in 8 studies,stressing the ethical obligations of AI developers and users in sports contexts.The thematic analysis revealed overlapping concerns,as some studies addressed multiple issues simultaneously.Conclusion:Future research should focus on developing ethical frameworks tailored to underrepresented sports contexts and creating global standards for AI regulation in sports.This includes investigating the implications of AI applications in amateur sports,enhancing diversity in AI training datasets,and exploring the integration of ethical AI practices across various sports governance structures.展开更多
This paper investigates the sliding-mode-based fixed-time distributed average tracking (DAT) problem for multiple Euler-Lagrange systems in the presence of external distur-bances. The primary objective is to devise co...This paper investigates the sliding-mode-based fixed-time distributed average tracking (DAT) problem for multiple Euler-Lagrange systems in the presence of external distur-bances. The primary objective is to devise controllers for each agent, enabling them to precisely track the average of multiple time-varying reference signals. By averaging these signals, we can mitigate the influence of errors and uncertainties arising dur-ing measurements, thereby enhancing the robustness and stabi-lity of the system. A distributed fixed-time average estimator is proposed to estimate the average value of global reference sig-nals utilizing local information and communication with neigh-bors. Subsequently, a fixed-time sliding mode controller is intro-duced incorporating a state-dependent sliding mode function coupled with a variable exponent coefficient to achieve dis-tributed average tracking of reference signals, and rigorous ana-lytical methods are employed to substantiate the fixed-time sta-bility. Finally, numerical simulation results are provided to vali-date the effectiveness of the proposed methodology, offering insights into its practical application and robust performance.展开更多
On May 19,2025,the groundbreaking ceremony for the Phase I of the nylon fiber project of Colorful Nylon Fiber Co.,Ltd.(a subsidiary of Eversun Corporation),was held in the Dat Do Industrial Park in Vung Tau Province,V...On May 19,2025,the groundbreaking ceremony for the Phase I of the nylon fiber project of Colorful Nylon Fiber Co.,Ltd.(a subsidiary of Eversun Corporation),was held in the Dat Do Industrial Park in Vung Tau Province,Vietnam.Representatives from the Eversun Corporation,the local government of Vietnam,the Vung Tau Industrial Park,and cooperative enterprises jointly witnessed this important moment.Jiangen Wang,General Manager of Eversun Corporation,Yuxin Chen,General Manager of Resultant Construction Co.,Ltd.,and Youtong Chen,Deputy Director of Dat Do Industrial Park Management Committee attended the ceremony.展开更多
Fundamental physics often confronts complex symbolic problems with few guiding exemplars or established principles.While artificial intelligence(AI)offers promise,its typical need for vast datasets to learn from hinde...Fundamental physics often confronts complex symbolic problems with few guiding exemplars or established principles.While artificial intelligence(AI)offers promise,its typical need for vast datasets to learn from hinders its use in these information-scarce frontiers.We introduce learning at criticality(LaC),a reinforcement learning scheme that tunes large language models(LLMs)to a sharp learning transition,addressing this information scarcity.At this transition,LLMs achieve peak generalization from minimal data,exemplified by 7-digit base-7 addition-a test of nontrivial arithmetic reasoning.To elucidate this peak,we analyze a minimal concept-network model designed to capture the essence of how LLMs might link tokens.Trained on a single exemplar,this model also undergoes a sharp learning transition.This transition exhibits hallmarks of a second-order phase transition,notably power-law distributed solution path lengths.At this critical point,the system maximizes a“critical thinking pattern”crucial for generalization,enabled by the underlying scale-free exploration.This suggests LLMs reach peak performance by operating at criticality,where such explorative dynamics enable the extraction of underlying operational rules.We demonstrate LaC in quantum field theory:an 8B-parameter LLM,tuned to its critical point by LaC using a few exemplars of symbolic Matsubara sums,solves unseen,higher-order problems,significantly outperforming far larger models.LaC thus leverages critical phenomena,a physical principle,to empower AI for complex,data-sparse challenges in fundamental physics.展开更多
文摘地球大气层外太阳光谱辐照度(extraterrestrial solar spectral irradiance,ESSI)数据是计算卫星传感器波段平均太阳辐照度(band mean solar irradiance,BMSI)的重要参数。为了探求利用何种来源的ESSI数据计算传感器BMSI更为准确,分别采用SBDART软件模拟的太阳光谱曲线数据、MODTRAN4.0 oldkur.dat文件数据、Thuillier太阳光谱曲线数据和WRC太阳光谱曲线数据计算了HJ-1A CCD1(B1—B4),CBERS-02 CCD(B1—B5),Landsat5TM(B1—B4)和ASTER(B1—B8)4种传感器的BMSI,并与传感器运营商公布的数据进行了比较。结果表明:利用SBDART和WRC太阳光谱曲线数据计算的结果误差较小;利用MODTRAN4.0 oldkur.dat数据计算的结果误差次之;利用Thuillier太阳光谱曲线的计算结果误差较大。
文摘目的总结1例Rh系统抗-c胎儿新生儿溶血病(hemolytic disease of the fetus and newborn,HDFN)的实验室检测并进行文献复习,探究抗-c HDFN的特点。方法采用血清学方法检测患儿及其母亲的ABO血型、Rh分型、直接抗人球蛋白试验(DAT)、不规则抗体及其效价,并对本实验室及国内外抗-c HDFN案例进行统计,分析比较抗-c与抗-D、抗-E引起HDFN的严重率。结果患儿血型为B、CcDee,DAT阳性,血清和放散液均检测出抗-c,血清抗体效价为4;母亲血型为AB、CCDee,DAT阴性,血清中检测出抗-c,效价为128。20例抗-c HDFN案例中DAT阳性有17例,输血或换血的有9例,占45%(9/20)。严重率为:抗-c 47.60%(10/21)、抗-D 47.60%(10/21)、抗-E 31.30%(5/16)。结论母亲妊娠和/或输血是产生抗-c等Rh系统同种抗体的主要原因。抗-c的预防管理应类同抗-D,对育龄女性输血要做好Rh血型5种抗原匹配输注以避免抗体产生,并重视孕妇产检Rh血型鉴定和抗体筛查鉴定,做到早发现、早干预、早治疗。
文摘谱聚类因其在建模数据间成对相似关系方面的优越性而广泛应用于无监督学习领域.然而,传统谱聚类方法通常依赖干净、结构一致的数据分布,在现实应用中面临常见的噪声样本时,性能显著下降.针对该问题,文中提出融合CLIP(Contrastive Language Image Pretraining)先验知识的谱聚类框架——基于知识重用的噪声环境谱聚类(Noise Spectral Clustering with Assistance of Knowledge Reuse,NSCR).该方法充分利用多模态神经网络在跨模态语义理解上的先验能力,构建基于知识重用的伪标签生成机制,通过多模型语义一致性判别机制与基于信息熵的不确定性建模机制识别高可信样本.同时引入归一化指数熵作为伪标签不确定性度量指标,从多模型输出中筛选语义一致、信息熵较低的样本,并生成伪标签,监督信号形式,引导聚类过程.此外,引入联合优化目标,扩展传统谱聚类方法,通过特征对齐与正则化平衡因子缓解伪标签监督与聚类目标之间的语义冲突.在多个公开数据集上的实验表明,NSCR在不同类型噪声干扰下的鲁棒性与泛化性良好.
文摘Background:Although there is growing evidence of the use of artificial intelligence(AI)techniques in sports,ethical issues surrounding AI use are being discussed at a minimal level.Thus,this systematic scoping review aimed to summarize the current ethical implications associated with using AI in sports.Methods:In this study,a total of 9 databases-MEDLINE/PubMed,Embase,Cochrane Library,ProQuest,EBSCOhost,IEEE Xplore,Web of Science,Scopus,and Google Scholar--were searched.The review protocol was registered(https://osfio/42a8q)before extracting data.The search yielded 397 studies,and 25 studies met the inclusion and exclusion criteria.Results:The 25 studies were categorized into 4 primary ethical concerns:fairness and bias,transparency and explainability,privacy and data ethics,and accountability in AI's application in sports.These categorizations were derived based on the systematic review ofethical issues highlighted across the selected studies.Fifteen studies delved into fairness and bias,focusing on how AI can perpetuate existing inequalities in sports.Thirteen studies addressed the lack of transparency,emphasizing the challenges in interpretability and trust in AI-driven decisions.Privacy and data ethics emerged as significant in22 studies,highlighting risks related to the misuse of athletes’sensitive data.Finally,account-ability was examined in 8 studies,stressing the ethical obligations of AI developers and users in sports contexts.The thematic analysis revealed overlapping concerns,as some studies addressed multiple issues simultaneously.Conclusion:Future research should focus on developing ethical frameworks tailored to underrepresented sports contexts and creating global standards for AI regulation in sports.This includes investigating the implications of AI applications in amateur sports,enhancing diversity in AI training datasets,and exploring the integration of ethical AI practices across various sports governance structures.
基金supported by the National Natural Science Foundation of China(61673130).
文摘This paper investigates the sliding-mode-based fixed-time distributed average tracking (DAT) problem for multiple Euler-Lagrange systems in the presence of external distur-bances. The primary objective is to devise controllers for each agent, enabling them to precisely track the average of multiple time-varying reference signals. By averaging these signals, we can mitigate the influence of errors and uncertainties arising dur-ing measurements, thereby enhancing the robustness and stabi-lity of the system. A distributed fixed-time average estimator is proposed to estimate the average value of global reference sig-nals utilizing local information and communication with neigh-bors. Subsequently, a fixed-time sliding mode controller is intro-duced incorporating a state-dependent sliding mode function coupled with a variable exponent coefficient to achieve dis-tributed average tracking of reference signals, and rigorous ana-lytical methods are employed to substantiate the fixed-time sta-bility. Finally, numerical simulation results are provided to vali-date the effectiveness of the proposed methodology, offering insights into its practical application and robust performance.
文摘On May 19,2025,the groundbreaking ceremony for the Phase I of the nylon fiber project of Colorful Nylon Fiber Co.,Ltd.(a subsidiary of Eversun Corporation),was held in the Dat Do Industrial Park in Vung Tau Province,Vietnam.Representatives from the Eversun Corporation,the local government of Vietnam,the Vung Tau Industrial Park,and cooperative enterprises jointly witnessed this important moment.Jiangen Wang,General Manager of Eversun Corporation,Yuxin Chen,General Manager of Resultant Construction Co.,Ltd.,and Youtong Chen,Deputy Director of Dat Do Industrial Park Management Committee attended the ceremony.
基金supported by the National Key Research and Development Program of China(Grant No.2024YFA1408604 for K.C.and X.C.)the National Natural Science Foundation of China(Grant Nos.12047503,12447103 for K.C.and X.C.,12325501 for P.Z.,and 12275263 for Y.D.and S.H.)+1 种基金the Innovation Program for Quantum Science and Technology(Grant No.2021ZD0301900 for Y.D.and S.H.)the Natural Science Foundation of Fujian Province of China(Grant No.2023J02032 for Y.D.and S.H.)。
文摘Fundamental physics often confronts complex symbolic problems with few guiding exemplars or established principles.While artificial intelligence(AI)offers promise,its typical need for vast datasets to learn from hinders its use in these information-scarce frontiers.We introduce learning at criticality(LaC),a reinforcement learning scheme that tunes large language models(LLMs)to a sharp learning transition,addressing this information scarcity.At this transition,LLMs achieve peak generalization from minimal data,exemplified by 7-digit base-7 addition-a test of nontrivial arithmetic reasoning.To elucidate this peak,we analyze a minimal concept-network model designed to capture the essence of how LLMs might link tokens.Trained on a single exemplar,this model also undergoes a sharp learning transition.This transition exhibits hallmarks of a second-order phase transition,notably power-law distributed solution path lengths.At this critical point,the system maximizes a“critical thinking pattern”crucial for generalization,enabled by the underlying scale-free exploration.This suggests LLMs reach peak performance by operating at criticality,where such explorative dynamics enable the extraction of underlying operational rules.We demonstrate LaC in quantum field theory:an 8B-parameter LLM,tuned to its critical point by LaC using a few exemplars of symbolic Matsubara sums,solves unseen,higher-order problems,significantly outperforming far larger models.LaC thus leverages critical phenomena,a physical principle,to empower AI for complex,data-sparse challenges in fundamental physics.
