首页期刊导航|研究（英文）
期刊信息/Journal information
研究（英文）
研究（英文）
研究（英文）/Journal ResearchCSCDCSTPCD北大核心SCI
正式出版
收录年代

    Parallel Driving with Big Models and Foundation Intelligence in Cyber-Physical-Social Spaces

    Xiao WangJun HuangYonglin TianChen Sun...
    1-17页
    查看更多>>摘要：Recent years have witnessed numerous technical breakthroughs in connected and autonomous vehicles(CAVs).On the one hand,these breakthroughs have significantly advanced the development of intelligent transportation systems(ITSs);on the other hand,these new traffic participants introduce more complex and uncertain elements to ITSs from the social space.Digital twins(DTs)provide real-time,data-driven,precise modeling for constructing the digital mapping of physical-world ITSs.Meanwhile,the metaverse integrates emerging technologies such as virtual reality/mixed reality,artificial intelligence,and DTs to model and explore how to realize improved sustainability,increased efficiency,and enhanced safety.More recently,as a leading effort toward general artificial intelligence,the concept of foundation model was proposed and has achieved significant success,showing great potential to lay the cornerstone for diverse artificial intelligence applications across different domains.In this article,we explore the big models embodied foundation intelligence for parallel driving in cyber-physical-social spaces,which integrate metaverse and DTs to construct a parallel training space for CAVs,and present a comprehensive elucidation of the crucial characteristics and operational mechanisms.Beyond providing the infrastructure and foundation intelligence of big models for parallel driving,this article also discusses future trends and potential research directions,and the"6S"goals of parallel driving.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Cross-Modal Graph Semantic Communication Assisted by Generative Al in the Metaverse for 6G

    Mingkai ChenMinghao LiuCongyan WangXingnuo Song...
    19-27页
    查看更多>>摘要：Recently,the development of the Metaverse has become a frontier spotlight,which is an important demonstration of the integration innovation of advanced technologies in the Internet.Moreover,artificial intelligence(AI)and 6G communications will be widely used in our daily lives.However,the effective interactions with the representations of multimodal data among users via 6G communications is the main challenge in the Metaverse.In this work,we introduce an intelligent cross-modal graph semantic communication approach based on generative Al and 3-dimensional(3D)point clouds to improve the diversity of multimodal representations in the Metaverse.Using a graph neural network,multimodal data can be recorded by key semantic features related to the real scenarios.Then,we compress the semantic features using a graph transformer encoder at the transmitter,which can extract the semantic representations through the cross-modal attention mechanisms.Next,we leverage a graph semantic validation mechanism to guarantee the exactness of the overall data at the receiver.Furthermore,we adopt generative Al to regenerate multimodal data in virtual scenarios.Simultaneously,a novel 3D generative reconstruction network is constructed from the 3D point clouds,which can transfer the data from images to 3D models,and we infer the multimodal data into the 3D models to increase realism in virtual scenarios.Finally,the experiment results demonstrate that cross-modal graph semantic communication,assisted by generative Al,has substantial potential for enhancing user interactions in the 6G communications and Metaverse.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Enhancing Offensive Language Detection with Data Augmentation and Knowledge Distillation

    Jiawen DengZhuang ChenHao SunZhexin Zhang...
    29-40页
    查看更多>>摘要：Offensive language detection has received important attention and plays a crucial role in promoting healthy communication on social platforms,as well as promoting the safe deployment of large language models.Training data is the basis for developing detectors;however,the available offense-related dataset in Chinese is severely limited in terms of data scale and coverage when compared to English resources.This significantly affects the accuracy of Chinese offensive language detectors in practical applications,especially when dealing with hard cases or out-of-domain samples.To alleviate the limitations posed by available datasets,we introduce AugCOLD(Augmented Chinese Offensive Language Dataset),a large-scale unsupervised dataset containing 1 million samples gathered by data crawling and model generation.Furthermore,we employ a multiteacher distillation framework to enhance detection performance with unsupervised data.That is,we build multiple teachers with publicly accessible datasets and use them to assign soft labels to AugCOLD.The soft labels serve as a bridge for knowledge to be distilled from both AugCOLD and multiteacher to the student network,i.e.,the final offensive detector.We conduct experiments on multiple public test sets and our well-designed hard tests,demonstrating that our proposal can effectively improve the generalization and robustness of the offensive language detector.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Multiplayer Reach-Avoid Differential Games in 3D Space Inspired by Harris'Hawks'Cooperative Hunting Tactics

    Wanying RuanHaibin DuanYongbin SunWanmai Yuan...
    41-54页
    查看更多>>摘要：This paper investigates a multiplayer reach-avoid differential game in 3-dimensional(3D)space,which involves multiple pursuers,multiple evaders,and a designated target region.The evaders aim to reach the target region,while the pursuers attempt to guard the target region by capturing the evaders.This class of research holds significant practical value.However,the complexity of the problem escalates substantially with the growing number of players,rendering its solution extremely challenging.In this paper,the multiplayer game is divided into many subgames considering the cooperation among pursuers,reducing the computational burden,and obtaining numerically tractable strategies for players.First,the Apollonius sphere,a fundamental geometric tool for analyzing the 3D differential game,is formulated,and its properties are proved.Based on this,the optimal interception point for the pursuer to capture the evader is derived and the winning conditions for the pursuer and evader are established.Then,based on the Apollonius sphere,the optimal state feedback strategies of players are designed,and simultaneously,the optimal one-to-one pairings are obtained.Meanwhile,the Value function of the multiplayer reach-avoid differential game is explicitly given and is proved to satisfy Hamilton-Jacobi-Isaacs(HJI)equation.Moreover,the matching algorithm for the case with pursuers outnumbered evaders is provided through constructing a weighted bipartite graph,and the cooperative tactics for multiple pursuers are proposed,inspired by the Harris'Hawks intelligent cooperative hunting tactics.Finally,numerical simulations are conducted to illustrate the effectiveness of the theoretical results for both cases where the number of adversary players is equal and unequal between the 2 groups.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Stochastic Computing Convolutional Neural Network Architecture Reinvented for Highly Efficient Artificial Intelligence Workload on Field-Programmable Gate Array

    Yang Yang LeeZaini Abdul HalimMohd Nadhir Ab WahabTarik Adnan Almohamad...
    55-79页
    查看更多>>摘要：Stochastic computing(SC)has a substantial amount of study on application-specific integrated circuit(ASIC)design for artificial intelligence(Al)edge computing,especially the convolutional neural network(CNN)algorithm.However,SC has little to no optimization on field-programmable gate array(FPGA).Scaling up the ASIC logic without FPGA-oriented designs is inefficient,while aggregating thousands of bitstreams is still challenging in the conventional SC.This research has reinvented several FPGA-efficient 8-bit SC CNN computing architectures,i.e.,SC multiplexer multiply-accumulate,multiply-accumulate function generator,and binary rectified linear unit,and successfully scaled and implemented a fully parallel CNN model on Kintex7 FPGA.The proposed SC hardware only compromises 0.14％accuracy compared to binary computing on the handwriting Modified National Institute of Standards and Technology classification task and achieved at least 99.72％energy saving per image feedforward and 31x more data throughput than modern hardware.Unique to SC,early decision termination pushed the performance baseline exponentially with minimum accuracy loss,making SC CNN extremely lucrative for Al edge computing but limited to classification tasks.The SC's inherent noise heavily penalizes CNN regression performance,rendering SC unsuitable for regression tasks.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Fast and Adaptive Multi-Agent Planning under Collaborative Temporal Logic Tasks via Poset Products

    Zesen LiuMeng GuoWeimin BaoZhongkui Li...
    81-94页
    查看更多>>摘要：Efficient coordination and planning is essential for large-scale multi-agent systems that collaborate in a shared dynamic environment.Heuristic search methods or learning-based approaches often lack the guarantee on correctness and performance.Moreover,when the collaborative tasks contain both spatial and temporal requirements,e.g.,as linear temporal logic(LTL)formulas,formal methods provide a verifiable framework for task planning.However,since the planning complexity grows exponentially with the number of agents and the length of the task formula,existing studies are mostly limited to small artificial cases.To address this issue,a new planning paradigm is proposed in this work for system-wide temporal task formulas that are released online and continually.It avoids two common bottlenecks in the traditional methods,i.e.,(a)the direct translation of the complete task formula to the associated Büchi automaton and(b)the synchronized product between the Büchi automaton and the transition models of all agents.Instead,an adaptive planning algorithm is proposed,which computes the product of relaxed partially ordered sets(R-posets)on-the-fly and assigns these subtasks to the agents subject to the ordering constraints.It is shown that the first valid plan can be derived with a polynomial time and memory complexity with respect to the system size and the formula length.Our method can take into account task formulas with a length of more than 400 and a fleet with more than 400 agents,while most existing methods fail at the formula length of 25 within a reasonable duration.The proposed method is validated on large fleets of service robots in both simulation and hardware experiments.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Run-and-Tumble Dynamics and Mechanotaxis Discovered in Microglial Migration

    Yiyu ZhangDa WeiXiaochen WangBoyi Wang...
    95-103页
    查看更多>>摘要：Microglia are resident macrophage cells in the central nervous system that search for pathogens or abnormal neural activities and migrate to resolve the issues.The effective search and targeted motion of macrophages mean dearly to maintaining a healthy brain,yet little is known about their migration dynamics.In this work,we study microglial motion with and without the presence of external mechanostimuli.We discover that the cells are promptly attracted by the applied forces(i.e.,mechanotaxis),which is a tactic behavior as yet unconfirmed in microglia.Meanwhile,in both the explorative and the targeted migration,microglia display dynamics that is strikingly analogous to bacterial run-and-tumble motion.A closer examination reveals that microglial run-and-tumble is more sophisticated,e.g.,they display a short-term memory when tumbling and rely on active steering during runs to achieve mechanotaxis,probably via the responses of mechanosensitive ion channels.These differences reflect the sharp contrast between microglia and bacteria cells(eukaryotes vs.prokaryotes)and their environments(compact tissue vs.fluid).Further analyses suggest that the reported migration dynamics has an optimal search efficiency and is shared among a subset of immune cells(human monocyte and macrophage).This work reveals a fruitful analogy between the locomotion of 2 remote systems and provides a framework for studying immune cells exploring complex environments.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Molecular Basis of KAT2A Selecting Acyl-CoA Cofactors for Histone Modifications

    Sha LiNan LiJie HeRunxin Zhou...
    105-109页
    查看更多>>摘要：Emerging discoveries about undocumented acyltransferase activities of known histone acetyltransferases(HATs)advance our understandings in the regulation of histone modifications.However,the molecular basis of HATs selecting acyl coenzyme A(acyl-CoA)substrates for histone modification is less known.We here report that lysine acetyltransferase 2A(KAT2A)as an illustrative instance of HATs can selectively utilize acetyl-CoA,propionyl-CoA,butyryl-CoA,and succiny l-CoA to directly deposit 18 histone acylation hallmarks in nucleosome.By analyzing the co-crystal structures of the catalytic domain of KAT2A in complex with acetyl-CoA,propionyl-CoA,butyryl-CoA,malonyl-CoA,succinyl-CoA,and glutary l-CoA,we conclude that the alternative substrate-binding pocket of KAT2A and the length and electrostatic features of the acyl chain cooperatively determine the selection of the acyl-CoA substrates by KAT2A.This study reveals the molecular basis underlying the pluripotency of HATs that selectively install acylation hallmarks in nucleosomes,which might serve as instrumental mechanism to precisely regulate histone acylation profiles in cells.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Carrier Dynamics Determines the Optimization Strategies of Perovskite LEDs and PVs

    Saixue WangYu CaoQiming PengWei Huang...
    111-114页
    查看更多>>摘要：Metal halide perovskites have advanced greatly in both light-emitting diodes(LEDs)and photovoltaics(PVs)through delicate device engineering.The optimization strategies of perovskite LEDs and PVs have been demonstrated to be quite different.Here,we show that this dissimilarity in device fabrications can be well understood based on the analysis of carrier dynamics in LEDs and PVs.
      原文链接:
    • NETL
    • NSTL
    • 万方数据

    Aerodynamic Super-Repellent Surfaces

    Fanfei YuJinlong YangRan TaoYao Tan...
    115-122页
    查看更多>>摘要：Repelling liquid drops from engineering surfaces has attracted great attention in a variety of applications.To achieve efficient liquid shedding,delicate surface textures are often introduced to sustain air pockets at the liquid-solid interface.However,those surfaces are prone to suffer from mechanical failure,which may bring reliability issues and thus limits their applications.Here,inspired by the aerodynamic Leidenfrost effect,we present that impacting drops are directionally repelled from smooth surfaces supplied with an exogenous air layer.Our theoretical analysis reveals that the synchronized nonwetting and oblique bouncing behavior is attributed to the aerodynamic force arising from the air layer.The versatility and practicability of our approach allow for drop repellency without the aid of any surface wettability treatment and also avoid the consideration of mechanical stability issues,which thereby provides a promising candidate for the applications that necessitate liquid shedding,e.g.,resolve the problem of tiny raindrop adhesion on the automobile side window during driving.
      原文链接:
    • NETL
    • NSTL
    • 万方数据