检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Chenglong Tang, Haoyu Yang, Xiyuan Xie, Yueqi Suh, Jina Sitaram, Sunayana Huang, Junming Xie, Yu Gong, Zhaoya Xie, Xing Wu, Fangzhao School of Urban Planning and Design Peking University Shenzhen Graduate School Shenzhen China Key Laboratory of Earth Surface System and Human-Earth Relations Ministry of Natural Resources of China Peking University Shenzhen Graduate School Shenzhen China School of Computer Science and Technology University of Science and Technology of China Hefei China School of Computer Science Wuhan University Wuhan China Hong Kong University of Science and Technology Hong Kong Microsoft Research Redmond United States Microsoft Research Bengaluru India Paul and Marcia Center on Contemporary China Princeton University Princeton United States Center for Social Research Guanghua School of Management Peking University Beijing China Microsoft Research Asia Beijing China

As large language models (LLMs) gradually become integral tools for problem solving in daily life worldwide, understanding linguistic inequality is becoming increasingly important. Existing research has primarily focused on static analyses that assess the disparities in the existing knowledge and capabilities of LLMs across languages. However, LLMs are continuously evolving, acquiring new knowledge to generate up-to-date, domain-specific responses. Investigating linguistic inequalities within this dynamic process is, therefore, also essential. In this paper, we explore inequalities in new knowledge learning by LLMs across different languages and four key dimensions: effectiveness, transferability, prioritization, and robustness. Through extensive experiments under two settings (in-context learning and fine-tuning) using both proprietary and open-source models, we demonstrate that low-resource languages consistently face disadvantages across all four dimensions. By shedding light on these disparities, we aim to raise awareness of linguistic inequities in LLMs’ new knowledge learning, fostering the development of more inclusive and equitable future LLMs. Copyright © 2025, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Generative API usage code recommendation with parameter concretization

引用

science China(Information sciences) 2019年第9期62卷 51-72页

作者： Chi CHEN Xin PENG Jun SUN Zhenchang XING Xin WANG Yifan ZHAO Hairui ZHANG Wenyun ZHAO School of Computer Science Fudan University Shanghai Key Laboratory of Data Science Fudan University Shanghai Institute of Intelligent Electronics & Systems Pillar of Information System Technology and Design Singapore University of Technology and Design Research School of Computer Science Australian National University

Many programming languages and development frameworks have extensive libraries(e.g., JDK and Android libraries) that ease the task of software engineering if used effectively. With numerous library classes and sometimes intricate API(application programming interface) usage constraints, programmers often have difficulty remembering the library APIs and/or using them correctly. This study addresses this problem by developing an engine called DeepAPIRec, which automatically recommends the API usage *** to the existing proposals, our approach distinguishes itself in two ways. First, it is based on a tree-based long short-term memory(LSTM) neural network inspired by recent developments in the machinelearning community. A tree-based LSTM neural network allows us to model and reason about variable-length,preceding and succeeding code contexts, and to make precise predictions. Second, we apply data-flow analysis to generate concrete parameters for the API usage code, which not only allows us to generate complete code recommendations but also improves the accuracy of the learning results according to the tree-based LSTM neural network. Our approach has been implemented for supporting Java programs. Our experimental studies on the JDK library show that at statement-level recommendations, DeepAPIRec can achieve a top-1 accuracy of about 37% and a top-5 accuracy of about 64%, which are significantly better than the existing approaches. Our user study further confirms that DeepAPIRec can help developers to complete a segment of code faster and more accurately as compared to IntelliJ IDEA.

关键词： code recommendation API deep learning data dependency parameter concretization

来源：评论

学校读者我要写书评

暂无评论

Distilling Autoregressive Models to Obtain High-Performance Non-Autoregressive Solvers for Vehicle Routing Problems with Faster Inference Speed

arXiv

引用

arXiv 2023年

作者： Xiao, Yubin Wang, Di Li, Boyang Wang, Mingzhao Wu, Xuan Zhou, Changliang Zhou, You Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University China Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly Nanyang Technological University Singapore WeBank-NTU Joint Research Institute on Fintech Nanyang Technological University Singapore School of Computer Science and Engineering Nanyang Technological University Singapore School of System Design and Intelligent Manufacturing Southern University of Science and Technology China

Neural construction models have shown promising performance for Vehicle Routing Problems (VRPs) by adopting either the Autoregressive (AR) or Non-Autoregressive (NAR) learning approach. While AR models produce high-quality solutions, they generally have a high inference latency due to their sequential generation nature. Conversely, NAR models generate solutions in parallel with a low inference latency but generally exhibit inferior performance. In this paper, we propose a generic Guided Non-Autoregressive Knowledge Distillation (GNARKD) method to obtain high-performance NAR models having a low inference latency. GNARKD removes the constraint of sequential generation in AR models while preserving the learned pivotal components in the network architecture to obtain the corresponding NAR models through knowledge distillation. We evaluate GNARKD by applying it to three widely adopted AR models to obtain NAR VRP solvers for both synthesized and real-world instances. The experimental results demonstrate that GNARKD significantly reduces the inference time (4∼5× faster) with acceptable performance drop (2∼3%). To the best of our knowledge, this study is first-of-its-kind to obtain NAR VRP solvers from AR ones through knowledge distillation. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Data-driven design of Customized Porous Lattice Sole Fabricated by Additive Manufacturing

引用

Procedia Manufacturing 2021年 53卷 318-326页

作者： Yunlong Tang Guoying Dong Yi Xiong Qiusen Wang Mechanical and Aerospace Engineering Department Monash University Melbourne 3168 Australia Materials Science and Engineering Department Monash University Melbourne 3168 Australia Digital Manufacturing and Design Center Singapore University of Technology and Design Singapore 487372 School of System Design and Intelligent Manufacturing Southern University of Science and Technology Shenzhen 518055 China Chongqing Gonggangzhihui Additive Manufacturing Technology Research Institute Co. LTD

Additive Manufacturing technologies enable the fabrication of structures with multiscale complexities without increasing their cost. To take advantage of the unique design freedom enabled by additive manufacturing processes, this paper reports a new design method that can generate a fully customized porous shoe sole. The proposed design method contains five major steps and can generate a fully customized shoe sole with customized features on both macro and mesoscale. Specifically, compared to a conventional flat shoe sole, the top surface of a customized sole can fully conform to the bottom surface of patient’s feet, which can significantly reduce the peak plantar pressure. In addition to that, the strut diameters of designed lattice structures can also be customized based on the proposed data-driven design optimization algorithm. By varying the struts’ diameters on the different regions of the designed lattice shoe sole, the peak plantar pressure can be further minimized. The case study provided in this paper shows the proposed method that can significantly improve the performance of customized shoe sole. This promising result indicates that additive manufacturing fabricated lattice shoe soles can be a potential solution to prevent or treat diabetic ulcers.

关键词： Lattice structures Peak plantar pressure Shoe sole Additive manufacturing

来源：评论

学校读者我要写书评

暂无评论

Numerical Investigation on Thermal Characteristics and Energy Efficiency of Lithium-Ion Battery Under Different Charging Stresses

SSRN

引用

SSRN 2024年

作者： Xie, Wenlong Gao, Xinlei Zhang, Zhengjie Tianjin Key Laboratory for Advanced Mechatronic System Design and Intelligent Control School of Mechanical Engineering Tianjin University of Technology Tianjin300384 China National Demonstration Center for Experimental Mechanical and Electrical Engineering Education Tianjin University of Technology China Department of Mechanical Engineering Imperial College London LondonSW7 2AZ United Kingdom School of Transportation Science and engineering Beihang University Beijing100191 China

The thermal characteristics of lithium-ion batteries during the charging process significantly impact their electrochemical performance and safety, particularly under high-rate or extreme temperature conditions. In this study, an electrochemical-thermal coupling model for a 2.5 Ah LiNi0.8Co0.1Mn0.1O2 18650 lithium-ion battery is proposed and experimentally validated. The heat generation and variation patterns of the electrodes at different charging rates and temperatures are evaluated, and the influence mechanisms of various electrode design parameters on charging energy efficiency are elucidated. The results reveal that high-rate charging increases the proportion of irreversible heat in the total heat generation, primarily due to the increase in ohmic heating at the positive electrode. Ambient temperature is less sensitive to heat generation than the C-rate, but low temperatures can also increase irreversible heat and are associated with an increase in positive polarization heat. Analyzing thermal characteristics at the electrode level can contribute to the optimization of battery design and the improvement of thermal safety. Electrode architectures, including electrode thickness, active material porosity, particle radius and negative/positive electrode capacity (N/P) ratio, are determined to analyze their effects on the heat generation of various components within the battery. Thick electrodes or active materials with high porosity induce significant ohmic heating and increase irreversible heat, with a slight effect on reversible heat, which is detrimental to energy efficiency. Large particle radii primarily increase reaction heat and polarization heat, and a high N/P ratio reduces the battery heat generation and is related to the reduction of negative reaction heat. Notably, variations in both two parameters have limited impact on to energy efficiency. Therefore, implementing appropriate thermal management across various charging scenarios is crucial for achievin

关键词： Lithium-ion batteries

来源：评论

学校读者我要写书评

暂无评论

A Train Cooperative Operation Optimization Method Considering Passenger Comfort based on Reinforcement Learning 11

A Train Cooperative Operation Optimization Method Considerin...

引用

11th IEEE Data Driven Control and Learning systems Conference, DDCLS 2022

作者： Wang, Xingguo Wu, Yue Huang, Deqing Zhu, Lei Lu, Zheng He, Yan Institute of System Science and Technology School of Electrical Engineering Southwest Jiaotong University Chengdu611756 China Southwest Jiaotong University Key Laboratory of Advanced Energy Traction Comprehensive Energy Saving Railway Industry Chengdu610031 China School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China School of Design Arts Southwest Jiaotong University Chengdu610031 China

ISBN: (数字)9781665496759

ISBN: (纸本)9781665496759

This paper mainly focuses on the high-speed train cooperative operation problem, considering passenger comfort. To solve this problem, a speed curve optimization method based on Reinforcement Learning algorithm is proposed. First, according to the train dynamics system, we build the speed curve optimization object. For realizing the train cooperative operation and increasing the passenger comfort, trains distance constraint and jerk rate constraint are fully considered in formulating reward function. And then, agent of Twin Delayed Deep Deterministic Policy Gradient (TD3) is established. Based on TD3 algorithm, two groups of Q-value networks and one group of action action networks have the same structure, different initial values and update rates respectively, which improves the stability and convergence speed of the algorithm. After the training. The Agent can generate a desirable speed curve of train based on constraints of vehicle output and jerk rate under cooperative operation. © 2022 IEEE.

关键词： Speed

来源：评论

学校读者我要写书评

暂无评论

Network Learning from Best-Response Dynamics in LQ Games

Network Learning from Best-Response Dynamics in LQ Games

引用

American Control Conference (ACC)

作者： Yijun Chen Kemi Ding Guodong Shi Australian Centre for Field Robotics the University of Sydney Sydney Australia School of System Design and Intelligent Manufacturing Shenzhen Key Laboratory of Biomimetic Robotics and Intelligent Systems and the Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Southern University of Science and Technology Shenzhen China

In this paper, we focus on network structure inference problem for linear-quadratic (LQ) games from best-response dynamics. An adversary is considered to have no knowledge of the game network structure but have the ability to observe all players’ best-response actions and manipulate some players’ actions. This work presents a comprehensive framework for network learning from best-response dynamics in LQ games. First of all, we establish theoretic results that characterize network structure identifiability and provide numerical examples to demonstrate the usefulness of our theoretic results. Next, in the face of the inherent stability and sparsity constraints for the game network structure, we propose an information-theoretic stable and sparse system identification algorithm for learning the network structure. Finally, the effectiveness of the proposed learning algorithm is tested. The connection between network structure inference problem and classical system identification theory is covered by our work, which advances the literature.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Link Budget Calculation for LoRa-band Remoting Sensing Satellite

Link Budget Calculation for LoRa-band Remoting Sensing Satel...

引用

IEEE International Symposium on Electromagnetic Compatibility (EMC-Beijing)

作者： Shurui Wang Xiang Zhang Xiaojun Hao Shuaige Gong Chong Li Qian Xu Lei Xing College of Electronic and Information Engineering Nanjing University of Aeronautics and Astronautics Nanjing China College of Mechanical Engineering (Aircraft design and engineering) Nanjing University of Science and Technology Nanjing China State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System (CEMEE) Luoyang China

Small satellites are increasingly attracting the attention of the industry for its many aspects of advantages. We proposed a communication subsystem for commercial remote sensing satellite. The interface between the communication system and other modules on the satellite and the communication protocol are briefly introduced. In order to make sure that the satellite-to-ground link is reliable, we calculate the link budget at 471 MHz for LEO small satellite. Additionally, the transmission power and reception sensitivity is 17 dBm and −123 dBm respectively. The result shows that the proposed subsystem is reliable in RF performance and can be used in other commercial satellite in the future.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Understanding the Limitations of Thick Electrodes on the Rate Capability of High-Energy Density Lithium-Ion Batteries

SSRN

引用

SSRN 2024年

作者： Xie, Wenlong Zhang, Zhengjie Gao, Xinlei Tianjin Key Laboratory for Advanced Mechatronic System Design and Intelligent Control School of Mechanical Engineering Tianjin University of Technology Tianjin300384 China National Demonstration Center for Experimental Mechanical and Electrical Engineering Education Tianjin University of Technology China School of Transportation Science and engineering Beihang University Beijing100191 China Department of Mechanical Engineering Imperial College London LondonSW7 2AZ United Kingdom

As a core component of transportation electrification, lithium-ion batteries need to address two critical challenges: achieving high energy density and enabling fast charging. However, changes in electrode thickness could result in a trade-off between these two characteristics. Herein, the impact of electrode thickness on the electrochemical performance, thermal behavior, as well as energy and power density of LiNi0.8Co0.1Mn0.1O2(NCM)/graphite batteries are investigated through mathematical modeling. Underutilization of active materials of the thick electrode is identified as the primary limitation on rate capability, which can be attributed to the slow solid-state diffusion and poor electrolyte transport within the electrode, especially the enlarged concentration gradients in the solid phase and the lithium depletion in the electrolyte phase during high-rate charging. Evolutions of local current density and heat generation rate along the electrode thickness are compared, identifying the underlying mechanism of inducing uneven reaction kinetics inside the battery. The trade-off between the energy and power density as a function of electrode thickness is investigated based on the limitations of charge transfer and mass transport. Furthermore, the rate-limiting factors under various application requirements are elucidated. This study will provide guidance for the electrode design of high-energy-density batteries with fast-charging capabilities. © 2024, The Authors. All rights reserved.

关键词： Lithium-ion batteries

来源：评论

学校读者我要写书评

暂无评论

Cooperative Integrated Sensing and Communication Networks: Analysis and Distributed design

arXiv

引用

arXiv 2024年

作者： Wang, Bowen Li, Hongyu Liu, Fan Cheng, Ziyang Shen, Shanpu School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China Department of Electrical & Electronic Engineering Imperial College London LondonSW7 2AZ United Kingdom School of System Design and Intelligent Manufacturing Southern University of Sciencen and Technology Shenzhen518055 China Department of Electrical Engineering & Electronics University of Liverpool LiverpoolL69 3GJ United Kingdom

This paper proposes a cooperative integrated sensing and communication network (Co-ISACNet) adopting hybrid beamforming (HBF) architecture, which improves both radar sensing and communication performance. The main contributions of this work are four-fold. First, we introduce a novel cooperative sensing method for the considered Co-ISACNet, followed by a comprehensive analysis of this method. This analysis mathematically verifies the benefits of Co-ISACNet and provides insightful design guidelines. Second, to show the benefits of Co-ISACNet, we propose to jointly design the HBF to maximize the network communication capacity while satisfying the constraint of beampattern similarity for radar sensing, which results in a highly dimensional and non-convex problem. Third, to facilitate the joint design, we propose a novel distributed optimization framework based on proximal gradient and alternating direction method of multipliers, namely PANDA. Fourth, we further adopt the proposed PANDA framework to solve the joint HBF design problem for the Co-ISACNet. By using the proposed PANDA framework, all access points (APs) optimize the HBF in parallel, where each AP only requires local channel state information and limited message exchange among the APs. Such framework reduces significantly the computational complexity and thus has pronounced benefits in practical scenarios. Simulation results verify the effectiveness of the proposed algorithm compared with the conventional centralized algorithm and show the remarkable performance improvement of radar sensing and communication by deploying Co-ISACNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Channel state information

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：