The next-generation RAN,known as Open Radio Access Network(ORAN),allows for several advantages,including cost-effectiveness,network flexibility,and interoperability.Now ORAN applications,utilising machine learning(ML)...The next-generation RAN,known as Open Radio Access Network(ORAN),allows for several advantages,including cost-effectiveness,network flexibility,and interoperability.Now ORAN applications,utilising machine learning(ML)and artificial intelligence(AI)techniques,have become standard practice.The need for Federated Learning(FL)for ML model training in ORAN environments is heightened by the modularised structure of the ORAN architecture and the shortcomings of conventional ML techniques.However,the traditional plaintext model update sharing of FL in multi-BS contexts is susceptible to privacy violations such as deep-leakage gradient assaults and inference.Therefore,this research presents a novel blockchain-assisted improved cryptographic privacy-preserving federated learning(BICPPFL)model,with the help of ORAN,to safely carry out federated learning and protect privacy.This model improves on the conventional masking technique for sharing model parameters by adding new characteristics.These features include the choice of distributed aggregators,validation for final model aggregation,and individual validation for BSs.To manage the security and privacy of FL processes,a combined homomorphic proxy-reencryption(HPReE)and lattice-cryptographic method(HPReEL)has been used.The upgraded delegated proof of stake(Up-DPoS)consensus protocol,which will provide quick validation of model exchanges and protect against malicious attacks,is employed for effective consensus across blockchain nodes.Without sacrificing performance metrics,the BICPPFL model strengthens privacy and adds security layers while facilitating the transfer of sensitive data across several BSs.The framework is deployed on top of a Hyperledger Fabric blockchain to evaluate its effectiveness.The experimental findings prove the reliability and privacy-preserving capability of the BICPPFL model.展开更多
This paper aims to discuss and analyze the distribution characteristics of the private placement among different regions and different industries by state-owned and private holding enterprises as well as the changed e...This paper aims to discuss and analyze the distribution characteristics of the private placement among different regions and different industries by state-owned and private holding enterprises as well as the changed earning capacities of the capital before and after the placement. Taking the private placement financing data of the listed companies during the period of 2006-2011 in China's A-share markets as the sample, this study adopts the panel data model to conduct a quantitative analysis of the industrial upgrading effect of the private placement by listed companies with different holding natures. The empirical results show that though the private placement by the state-owned and private holding enterprises has not significantly improved the Clark index in any region, the private placement by private holding companies has significantly improved the growth rate of the tertiary industry in the eastern region. While reducing the proportion of the tertiary industry in GDP in the central and western regions, it has also improved the growth rate of the secondary industry and its proportion in GDP in that region to a larger extent. Yet the private placement of state-owned enterprises has no significant effect on the industrial upgrading path in any region except that it helped to raise the proportion of the secondary industry in GDP a little in the eastern region.展开更多
According to Bloomberg on November 12, China Petrochemical Corp., Asia’s biggest ref iner, agreed to buy a 30 percent stake in Galp Energia SGPS SA (GALP)’s Brazilian unit, its second invest- ment in offshore oil fi...According to Bloomberg on November 12, China Petrochemical Corp., Asia’s biggest ref iner, agreed to buy a 30 percent stake in Galp Energia SGPS SA (GALP)’s Brazilian unit, its second invest- ment in offshore oil fields in Latin America’s largest economy in as many years.展开更多
BP,the UK energy group,is planning to sell one of its biggest Chinese investments,by disposing of its 50percent stake in the SECCO petrochemicals plant near Shanghai.BP is the latest western oil company to curtail act...BP,the UK energy group,is planning to sell one of its biggest Chinese investments,by disposing of its 50percent stake in the SECCO petrochemicals plant near Shanghai.BP is the latest western oil company to curtail activity in China,as energy groups reel from low crude and petrochemical prices.China’s slow liberalization展开更多
How Egypt can attempt to revert back to its former glory OPTIMISTS who entertained thoughts that Egypt would revert to its former state of peace and tranquility after parliamentary elections must have been badly jolte...How Egypt can attempt to revert back to its former glory OPTIMISTS who entertained thoughts that Egypt would revert to its former state of peace and tranquility after parliamentary elections must have been badly jolted in early February.This is when a fairly innocuous event,a football match between two traditionally rival teams,ended with the death of over 70 people.How a sporting event can lead to such sectarian-motivated carnage underpins the deep-seated animosities unleashed after the January 25,2011 revo-展开更多
Background:Large language models(LLMs)have shown promise in educational applications,but their performance on high-stakes admissions tests,such as the Dental Admission Test(DAT),remains unclear.Understanding the capab...Background:Large language models(LLMs)have shown promise in educational applications,but their performance on high-stakes admissions tests,such as the Dental Admission Test(DAT),remains unclear.Understanding the capabilities and limitations of these models is critical for determining their suitability in test preparation.Methods:This study evaluated the ability of 16 LLMs,including general-purpose models(e.g.,GPT-3.5,GPT-4,GPT-4o,GPT-o1,Google’s Bard,mistral-large,and Claude),domain-specific finetuned models(e.g.,DentalGPT,MedGPT,and BioGPT),and open-source models(e.g.,Llama2-7B,Llama2-13B,Llama2-70B,Llama3-8B,and Llama3-70B),to answer questions from a sample DAT.Quantitative analysis was performed to assess model accuracy in different sections,and qualitative thematic analysis by subject matter experts examined specific challenges encountered by the models.Results:GPT-4o and GPT-o1 outperformed others in text-based questions assessing knowledge and comprehension,with GPT-o1 achieving perfect scores in the natural sciences(NS)and reading comprehension(RC)sections.Open-source models such as Llama3-70B also performed competitively in RC tasks.However,all models,including GPT-4o,struggled substantially with perceptual ability(PA)items,highlighting a persistent limitation in handling image-based tasks requiring visual-spatial reasoning.Fine-tuned medical models(e.g.,DentalGPT,MedGPT,and BioGPT)demonstrated moderate success in text-based tasks but underperformed in areas requiring critical thinking and reasoning.Thematic analysis identified key challenges,including difficulties with stepwise problem-solving,transferring knowledge,comprehending intricate questions,and hallucinations,particularly on advanced items.Conclusions:While LLMs show potential for reinforcing factual knowledge and supporting learners,their limitations in handling higherorder cognitive tasks and image-based reasoning underscore the need for judicious integration with instructor-led guidance and targeted practice.This study provides valuable insights into the capabilities and limitations of current LLMs in preparing prospective dental students and highlights pathways for future innovations to improve performance across all cognitive skills assessed by the DAT.展开更多
An idealized artificial general intelligence(AGI)system is expected to be smart enough to balance competing needs for caution and urgency via the management of limited time resources.However,mainstream Large Language ...An idealized artificial general intelligence(AGI)system is expected to be smart enough to balance competing needs for caution and urgency via the management of limited time resources.However,mainstream Large Language Models(LLMs),due to its working principle requiring operating resources of a tremendous magnitude,cannot be sensitive to the limitation of time resources,hence,it cannot adjust its time-distributions vis-à-vis the combinations of different internal and external conditions.Hence,a new AGI approach to the time-management is needed.By integrating the psychological notion of Need-For-Closure(NFC)and Nagel’s(2008)intellectual invariantism as well as Stanley’s(2005)stake-based epistemology,we will present a unified and computable framework for time-management,namely,a framework intended to handle both the system’s own estimation of the complexity of the task and its perception of the degree of stakes triggered by the task.In addition,in ourmodel,the computable notion of“budget deficit”plays a pivotal role in explaining how the NFC-value is fine-tuned to guide epistemic shifts.展开更多
先来看摘自一篇英语散文中的段落：地点哈瓦那,一个匈牙利人与倒卖雪茄的黑人烟贩子讨价还价。＂You’re crazy!＂cries the Hungarian in slightly accented English,taking one of the boxes from the trunk and waving it in Howard ...先来看摘自一篇英语散文中的段落：地点哈瓦那,一个匈牙利人与倒卖雪茄的黑人烟贩子讨价还价。＂You’re crazy!＂cries the Hungarian in slightly accented English,taking one of the boxes from the trunk and waving it in Howard Bingham’s face.＂These are Cohiba Esplendidos!The best in the world!You will pay one thousand dollars for a box like this in the States.＂展开更多
In the blockchain,the consensus mechanism plays a key role in maintaining the security and legitimation of contents recorded in the blocks.Various blockchain consensus mechanisms have been proposed.However,there is no...In the blockchain,the consensus mechanism plays a key role in maintaining the security and legitimation of contents recorded in the blocks.Various blockchain consensus mechanisms have been proposed.However,there is no technical analysis and comparison as a guideline to determine which type of consensus mechanism should be adopted in a specific scenario/application.To this end,this work investigates three mainstream consensus mechanisms in the blockchain,namely,Proof of Work(PoW),Proof of Stake(PoS),and Direct Acyclic Graph(DAG),and identifies their performances in terms of the average time to generate a new block,the confirmation delay,the Transaction Per Second(TPS)and the confirmation failure probability.The results show that the consensus process is affected by both network resource(computation power/coin age,buffer size)and network load conditions.In addition,it shows that PoW and PoS are more sensitive to the change of network resource while DAG is more sensitive to network load conditions.展开更多
文摘The next-generation RAN,known as Open Radio Access Network(ORAN),allows for several advantages,including cost-effectiveness,network flexibility,and interoperability.Now ORAN applications,utilising machine learning(ML)and artificial intelligence(AI)techniques,have become standard practice.The need for Federated Learning(FL)for ML model training in ORAN environments is heightened by the modularised structure of the ORAN architecture and the shortcomings of conventional ML techniques.However,the traditional plaintext model update sharing of FL in multi-BS contexts is susceptible to privacy violations such as deep-leakage gradient assaults and inference.Therefore,this research presents a novel blockchain-assisted improved cryptographic privacy-preserving federated learning(BICPPFL)model,with the help of ORAN,to safely carry out federated learning and protect privacy.This model improves on the conventional masking technique for sharing model parameters by adding new characteristics.These features include the choice of distributed aggregators,validation for final model aggregation,and individual validation for BSs.To manage the security and privacy of FL processes,a combined homomorphic proxy-reencryption(HPReE)and lattice-cryptographic method(HPReEL)has been used.The upgraded delegated proof of stake(Up-DPoS)consensus protocol,which will provide quick validation of model exchanges and protect against malicious attacks,is employed for effective consensus across blockchain nodes.Without sacrificing performance metrics,the BICPPFL model strengthens privacy and adds security layers while facilitating the transfer of sensitive data across several BSs.The framework is deployed on top of a Hyperledger Fabric blockchain to evaluate its effectiveness.The experimental findings prove the reliability and privacy-preserving capability of the BICPPFL model.
文摘This paper aims to discuss and analyze the distribution characteristics of the private placement among different regions and different industries by state-owned and private holding enterprises as well as the changed earning capacities of the capital before and after the placement. Taking the private placement financing data of the listed companies during the period of 2006-2011 in China's A-share markets as the sample, this study adopts the panel data model to conduct a quantitative analysis of the industrial upgrading effect of the private placement by listed companies with different holding natures. The empirical results show that though the private placement by the state-owned and private holding enterprises has not significantly improved the Clark index in any region, the private placement by private holding companies has significantly improved the growth rate of the tertiary industry in the eastern region. While reducing the proportion of the tertiary industry in GDP in the central and western regions, it has also improved the growth rate of the secondary industry and its proportion in GDP in that region to a larger extent. Yet the private placement of state-owned enterprises has no significant effect on the industrial upgrading path in any region except that it helped to raise the proportion of the secondary industry in GDP a little in the eastern region.
文摘According to Bloomberg on November 12, China Petrochemical Corp., Asia’s biggest ref iner, agreed to buy a 30 percent stake in Galp Energia SGPS SA (GALP)’s Brazilian unit, its second invest- ment in offshore oil fields in Latin America’s largest economy in as many years.
文摘BP,the UK energy group,is planning to sell one of its biggest Chinese investments,by disposing of its 50percent stake in the SECCO petrochemicals plant near Shanghai.BP is the latest western oil company to curtail activity in China,as energy groups reel from low crude and petrochemical prices.China’s slow liberalization
文摘How Egypt can attempt to revert back to its former glory OPTIMISTS who entertained thoughts that Egypt would revert to its former state of peace and tranquility after parliamentary elections must have been badly jolted in early February.This is when a fairly innocuous event,a football match between two traditionally rival teams,ended with the death of over 70 people.How a sporting event can lead to such sectarian-motivated carnage underpins the deep-seated animosities unleashed after the January 25,2011 revo-
基金partially supported by the National Institutes of Health’s National Center for Complementary and Integrative Health under grant number R01AT009457National Institute on Aging under grant number R01AG078154National Cancer Institute under grant number R01CA287413.
文摘Background:Large language models(LLMs)have shown promise in educational applications,but their performance on high-stakes admissions tests,such as the Dental Admission Test(DAT),remains unclear.Understanding the capabilities and limitations of these models is critical for determining their suitability in test preparation.Methods:This study evaluated the ability of 16 LLMs,including general-purpose models(e.g.,GPT-3.5,GPT-4,GPT-4o,GPT-o1,Google’s Bard,mistral-large,and Claude),domain-specific finetuned models(e.g.,DentalGPT,MedGPT,and BioGPT),and open-source models(e.g.,Llama2-7B,Llama2-13B,Llama2-70B,Llama3-8B,and Llama3-70B),to answer questions from a sample DAT.Quantitative analysis was performed to assess model accuracy in different sections,and qualitative thematic analysis by subject matter experts examined specific challenges encountered by the models.Results:GPT-4o and GPT-o1 outperformed others in text-based questions assessing knowledge and comprehension,with GPT-o1 achieving perfect scores in the natural sciences(NS)and reading comprehension(RC)sections.Open-source models such as Llama3-70B also performed competitively in RC tasks.However,all models,including GPT-4o,struggled substantially with perceptual ability(PA)items,highlighting a persistent limitation in handling image-based tasks requiring visual-spatial reasoning.Fine-tuned medical models(e.g.,DentalGPT,MedGPT,and BioGPT)demonstrated moderate success in text-based tasks but underperformed in areas requiring critical thinking and reasoning.Thematic analysis identified key challenges,including difficulties with stepwise problem-solving,transferring knowledge,comprehending intricate questions,and hallucinations,particularly on advanced items.Conclusions:While LLMs show potential for reinforcing factual knowledge and supporting learners,their limitations in handling higherorder cognitive tasks and image-based reasoning underscore the need for judicious integration with instructor-led guidance and targeted practice.This study provides valuable insights into the capabilities and limitations of current LLMs in preparing prospective dental students and highlights pathways for future innovations to improve performance across all cognitive skills assessed by the DAT.
文摘An idealized artificial general intelligence(AGI)system is expected to be smart enough to balance competing needs for caution and urgency via the management of limited time resources.However,mainstream Large Language Models(LLMs),due to its working principle requiring operating resources of a tremendous magnitude,cannot be sensitive to the limitation of time resources,hence,it cannot adjust its time-distributions vis-à-vis the combinations of different internal and external conditions.Hence,a new AGI approach to the time-management is needed.By integrating the psychological notion of Need-For-Closure(NFC)and Nagel’s(2008)intellectual invariantism as well as Stanley’s(2005)stake-based epistemology,we will present a unified and computable framework for time-management,namely,a framework intended to handle both the system’s own estimation of the complexity of the task and its perception of the degree of stakes triggered by the task.In addition,in ourmodel,the computable notion of“budget deficit”plays a pivotal role in explaining how the NFC-value is fine-tuned to guide epistemic shifts.
文摘先来看摘自一篇英语散文中的段落：地点哈瓦那,一个匈牙利人与倒卖雪茄的黑人烟贩子讨价还价。＂You’re crazy!＂cries the Hungarian in slightly accented English,taking one of the boxes from the trunk and waving it in Howard Bingham’s face.＂These are Cohiba Esplendidos!The best in the world!You will pay one thousand dollars for a box like this in the States.＂
基金the National Natural Science Foundation of China under Grant 61701059,Grant 61941114,and Grant 61831002,in part by the Fundamental Research Funds for the Central Universities of New TeachersProject,in part by the Chongqing Technological Innovation and Application Development Projects under Grant cstc2019jscx-msxm1322,and in part by the Eighteentg Open Foundation of State Key Lab of Integrated Services Networks of Xidian University under Grant ISN20-05.
文摘In the blockchain,the consensus mechanism plays a key role in maintaining the security and legitimation of contents recorded in the blocks.Various blockchain consensus mechanisms have been proposed.However,there is no technical analysis and comparison as a guideline to determine which type of consensus mechanism should be adopted in a specific scenario/application.To this end,this work investigates three mainstream consensus mechanisms in the blockchain,namely,Proof of Work(PoW),Proof of Stake(PoS),and Direct Acyclic Graph(DAG),and identifies their performances in terms of the average time to generate a new block,the confirmation delay,the Transaction Per Second(TPS)and the confirmation failure probability.The results show that the consensus process is affected by both network resource(computation power/coin age,buffer size)and network load conditions.In addition,it shows that PoW and PoS are more sensitive to the change of network resource while DAG is more sensitive to network load conditions.
