Adaptor signature,a new primitive that alleviates the scalability issue of blockchain to some extent,has been widely adopted in the off-chain payment channel and atomic swap.As an extension of standard digital signatu...Adaptor signature,a new primitive that alleviates the scalability issue of blockchain to some extent,has been widely adopted in the off-chain payment channel and atomic swap.As an extension of standard digital signature,adaptor signature can bind the release of a complete digital signature with the exchange of a secret value.Existing constructions of adaptor signatures are mainly based on Schnorr or ECDSA signature algorithms,which suffer low signing efficiency and long signature length.In this paper,to address these issues,we propose a new construction of adaptor signature using randomized EdDSA,which has Schnorr-like structure with higher signing efficiency and shorter signature length.We prove the required security properties,including unforgeability,witness extractability and pre-signature adaptability,of the new adaptor signature scheme in the random oracle model.We conduct a comparative analysis with an ECDSA-based adaptor signature scheme to demonstrate the effectiveness and feasibility of our new proposal.展开更多
Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple modalities.Despite these achievements,LLMs have in...Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple modalities.Despite these achievements,LLMs have inherent limitations including outdated information,hallucinations,inefficiency,lack of interpretability,and challenges in domain-specific accuracy.To address these issues,this survey explores three promising directions in the post-LLM era:knowledge empowerment,model collaboration,and model co-evolution.First,we examine methods of integrating external knowledge into LLMs to enhance factual accuracy,reasoning capabilities,and interpretability,including incorporating knowledge into training objectives,instruction tuning,retrieval-augmented inference,and knowledge prompting.Second,we discuss model collaboration strategies that leverage the complementary strengths of LLMs and smaller models to improve efficiency and domain-specific performance through techniques such as model merging,functional model collaboration,and knowledge injection.Third,we delve into model co-evolution,in which multiple models collaboratively evolve by sharing knowledge,parameters,and learning strategies to adapt to dynamic environments and tasks,thereby enhancing their adaptability and continual learning.We illustrate how the integration of these techniques advances AI capabilities in science,engineering,and society—particularly in hypothesis development,problem formulation,problem-solving,and interpretability across various domains.We conclude by outlining future pathways for further advancement and applications.展开更多
An oasis is a unique natural landscape in arid and semi-arid areas,significant for regulating regional microclimates and hydrological processes in deserts.However,little is known regarding the response of natural oasi...An oasis is a unique natural landscape in arid and semi-arid areas,significant for regulating regional microclimates and hydrological processes in deserts.However,little is known regarding the response of natural oasis plants communities to various environmental factors.Nineteen sample plots(50 m×50 m)were selected in the Daliyabuyi Oasis in the Taklimakan Desert hinterland based on the location of groundwater monitoring wells and 76 vegetation quadrats(25 m×25 m)were established.A two-way indicator species analysis,Mantel test,detrended correspondence analysis,canonical correspondence analysis(CCA),and hierarchical partitioning were used to provide an in-depth analysis of community classification,species composition,and environmental interpretation of the oasis.A generalized linear model was used to verify the results which showed that the current oasis community could be divided into four types according to the dominant species,which is controlled by soil moisture.Measurement of species composition and distribution of communities showed significant differences between species diversity of individual community types.Variations in groundwater depth affects patterns of species diversity which is sensitive to richness,while the degree of surface water disturbance affects the pattern of species evenness.Moreover,the CCA ordination map showed that community distribution and diversity characteristics have their own preferences in habitat gradients.The study concluded that the species dominance of the community and the composition and distribution are not dominated by a single factor.There are differences in the scale and effect of different water resource types in maintaining community characteristics.展开更多
Electromagnetic metasurfaces exhibit considerable potential for generating high-purity vortex beams and enabling high-resolution imaging and information encryption.However,traditional GHz devices face challenges,inclu...Electromagnetic metasurfaces exhibit considerable potential for generating high-purity vortex beams and enabling high-resolution imaging and information encryption.However,traditional GHz devices face challenges,including reduced efficiency due to bulky size and material losses.Herein,we designed a multilayer structure and demonstrated through simulations that this configuration served as an efficient transmissive meta-atom.We designed arrays in multiple sizes and finally determined that the optimal minimal unit was the meta-quaternion vortex array,which was subsequently used as the pixel basis for the target image.A digitally patterned GHz metadevice was fabricated and experimentally characterized with right-handed circularly polarized(RCP)light.The experimental results were in excellent agreement with the simulations.We combined the classical nine-grid encryption method(Luoshu)with metasurfaces and introduced the weighted superposition computation technique(WeightLock)to achieve multilayer encryption of target characters.Our research offered novel strategies for the next-generation 5G/6G communication systems,radar detection,and information encryption fields,demonstrating broad application prospects in intelligent communication and advanced radar technologies.展开更多
基金supported by the National Key R&D Program of China(2022YFB2701500)the National Natural Science Foundation of China(62272385,62311540156)+2 种基金Shaanxi Distinguished Youth Project(2022JC-47)the Key Research and Development Program of Shaanxi(2021ZDLGY06-04)Major Program of Shandong Provincial Natural Science Foundation for the Fundamental Research(ZR2022ZD03).
文摘Adaptor signature,a new primitive that alleviates the scalability issue of blockchain to some extent,has been widely adopted in the off-chain payment channel and atomic swap.As an extension of standard digital signature,adaptor signature can bind the release of a complete digital signature with the exchange of a secret value.Existing constructions of adaptor signatures are mainly based on Schnorr or ECDSA signature algorithms,which suffer low signing efficiency and long signature length.In this paper,to address these issues,we propose a new construction of adaptor signature using randomized EdDSA,which has Schnorr-like structure with higher signing efficiency and shorter signature length.We prove the required security properties,including unforgeability,witness extractability and pre-signature adaptability,of the new adaptor signature scheme in the random oracle model.We conduct a comparative analysis with an ECDSA-based adaptor signature scheme to demonstrate the effectiveness and feasibility of our new proposal.
基金supported in part by National Natural Science Foundation of China(62441605)。
文摘Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple modalities.Despite these achievements,LLMs have inherent limitations including outdated information,hallucinations,inefficiency,lack of interpretability,and challenges in domain-specific accuracy.To address these issues,this survey explores three promising directions in the post-LLM era:knowledge empowerment,model collaboration,and model co-evolution.First,we examine methods of integrating external knowledge into LLMs to enhance factual accuracy,reasoning capabilities,and interpretability,including incorporating knowledge into training objectives,instruction tuning,retrieval-augmented inference,and knowledge prompting.Second,we discuss model collaboration strategies that leverage the complementary strengths of LLMs and smaller models to improve efficiency and domain-specific performance through techniques such as model merging,functional model collaboration,and knowledge injection.Third,we delve into model co-evolution,in which multiple models collaboratively evolve by sharing knowledge,parameters,and learning strategies to adapt to dynamic environments and tasks,thereby enhancing their adaptability and continual learning.We illustrate how the integration of these techniques advances AI capabilities in science,engineering,and society—particularly in hypothesis development,problem formulation,problem-solving,and interpretability across various domains.We conclude by outlining future pathways for further advancement and applications.
基金supported by grants from the National Natural Science Foundation of the China Joint Key Program(No.U1703237)Postgraduate Research Innovation Project in the Autonomous Region(No.XJ2022G017)the National Natural Science Foundation of China Regional Program(No.32160260)。
文摘An oasis is a unique natural landscape in arid and semi-arid areas,significant for regulating regional microclimates and hydrological processes in deserts.However,little is known regarding the response of natural oasis plants communities to various environmental factors.Nineteen sample plots(50 m×50 m)were selected in the Daliyabuyi Oasis in the Taklimakan Desert hinterland based on the location of groundwater monitoring wells and 76 vegetation quadrats(25 m×25 m)were established.A two-way indicator species analysis,Mantel test,detrended correspondence analysis,canonical correspondence analysis(CCA),and hierarchical partitioning were used to provide an in-depth analysis of community classification,species composition,and environmental interpretation of the oasis.A generalized linear model was used to verify the results which showed that the current oasis community could be divided into four types according to the dominant species,which is controlled by soil moisture.Measurement of species composition and distribution of communities showed significant differences between species diversity of individual community types.Variations in groundwater depth affects patterns of species diversity which is sensitive to richness,while the degree of surface water disturbance affects the pattern of species evenness.Moreover,the CCA ordination map showed that community distribution and diversity characteristics have their own preferences in habitat gradients.The study concluded that the species dominance of the community and the composition and distribution are not dominated by a single factor.There are differences in the scale and effect of different water resource types in maintaining community characteristics.
基金National Key Research and Development Program of China(2024YFE107800)National Natural Science Foundation of China(62205127)+1 种基金Basic Research Program of Jiangsu(BK20231493)Key Laboratory of Advanced Optical Manufacturing Technologies of Jiangsu Province(KJS2272)。
文摘Electromagnetic metasurfaces exhibit considerable potential for generating high-purity vortex beams and enabling high-resolution imaging and information encryption.However,traditional GHz devices face challenges,including reduced efficiency due to bulky size and material losses.Herein,we designed a multilayer structure and demonstrated through simulations that this configuration served as an efficient transmissive meta-atom.We designed arrays in multiple sizes and finally determined that the optimal minimal unit was the meta-quaternion vortex array,which was subsequently used as the pixel basis for the target image.A digitally patterned GHz metadevice was fabricated and experimentally characterized with right-handed circularly polarized(RCP)light.The experimental results were in excellent agreement with the simulations.We combined the classical nine-grid encryption method(Luoshu)with metasurfaces and introduced the weighted superposition computation technique(WeightLock)to achieve multilayer encryption of target characters.Our research offered novel strategies for the next-generation 5G/6G communication systems,radar detection,and information encryption fields,demonstrating broad application prospects in intelligent communication and advanced radar technologies.
