检索结果-内蒙古大学图书馆

International Symposium on computers in Education (SIIE)

作者： Katerina Zdravkova Fisnik Dalipi Fredrik Ahlgren Faculty of Computer Science and Engineering Ss. Cyril and Methodius University Skopje N. Macedonia Department of Informatics Faculty of Technology Linnaeus University Kalmar Sweden Department of Computer Science and Media Technology Faculty of Technology Linnaeus University Kalmar Sweden

Large language models (LLMs) are being criticized for copyright infringement, inadvertent bias in training data, a danger to human innovation, the possibility of distributing incorrect or misleading information, and prejudice. Due to their popularity among students, the introduction of many comparable apps, and the inability to resist unfair and fraudulent student usage, their educational use needs to be adapted and harmonized. The incorporation of LLMs should be defined not only by pedagogues and educational institutions, but also by students who will actively utilize them to learn and prepare assignments. In order to find out what students from two universities think and suggest about LLMs use in education, they were asked to give their contribution by answering the survey that was conducted at the beginning of the spring semester of academic 2022/23. Their feedback was quantitatively and qualitatively analyzed, showing in a better light what students think about LLMs and how and why they would use them. Based on the analysis, the authors propose an original strategy for integrating LLMs into education. The proposed approach is also adapted for those students who are not interested in using LLMs and for those who prefer the hybrid mode by combining their own research with LLMs generated recommendations. The authors expect that by implementing the proposed strategy, schools will benefit from a better education in which research, creativity, academic honesty, recognition of false information, and the ability to improve knowledge will prevail.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models

arXiv

引用

arXiv 2025年

作者： Lai, Yuxiang Zhong, Jike Li, Ming Zhao, Shitian Yang, Xiaofeng Li, Yuheng Psounis, Konstantinos Department of Computer Science and Informatics Emory University AtlantaGA30322 United States Department of Computer Science Department of Electrical and Computer Engineering University of Southern California Los AngelesCA90089 United States Department of Computer Science University of Tokyo Tokyo113-8654 Japan Department of Computer Science Johns Hopkins University BaltimoreMD21218 United States Department of Biomedical Engineering Georgia Institute of Technology Emory University AtlantaGA30332 United States Department of Radiation Oncology Winship Cancer Institute Emory University AtlantaGA30322 United States

Vision-language models (VLMs) have achieved impressive progress in natural image reasoning, yet their potential in medical imaging remains underexplored. Medical vision-language tasks demand precise understanding and clinically coherent answers, which are difficult to achieve due to the complexity of medical data and the scarcity of high-quality expert annotations. These challenges limit the effectiveness of conventional supervised fine-tuning (SFT) and Chain-of-Thought (CoT) strategies that work well in general domains. To address these challenges, we propose Med-R1, a reinforcement learning (RL)-enhanced vision-language model designed to improve generalization and reliability in medical reasoning. Built on the DeepSeek strategy, Med-R1 adopts Group Relative Policy Optimization (GRPO) to encourage reward-guided learning beyond static annotations. We comprehensively evaluate Med-R1 across eight distinct medical imaging modalities. Med-R1 achieves a 29.94% improvement in average accuracy over its base model Qwen2-VL-2B, and even outperforms Qwen2-VL-72B—a model with 36× more parameters. To assess cross-task generalization, we further evaluate Med-R1 on five question types. Med-R1 outperforms Qwen2-VL-2B by 32.06% in question-type generalization, also surpassing Qwen2-VL-72B. We further explore the thinking process in Med-R1, a crucial component for the success of Deepseek-R1. Our results show that omitting intermediate rationales (No-Thinking-Med-R1) not only improves in-domain and cross-domain generalization with less training, but also challenges the assumption that more reasoning always helps. These findings suggest that in medical VQA, it is not reasoning itself, but its quality and domain alignment, that determine effectiveness. Together, these results highlight that RL improves medical reasoning and generalization, enabling efficient and reliable VLMs for real-world deployment. © 2025, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Tensorized Incomplete Multi-View Kernel Subspace Clustering

SSRN

引用

SSRN 2023年

作者： Zhang, Guang-Yu Huang, Dong Wang, Chang-Dong College of Mathematics and Informatics South China Agricultural University China School of Computer Science and Engineering Sun Yat-sen University China Guangdong Key Laboratory of Information Security Technology China

Recently considerable advances have been achieved in the incomplete multi-view clustering (IMC) research. However, the current IMC works are still faced with three challenging issues. First, they mostly lack the ability to recover the nonlinear subspace structure hidden in the multiple kernel spaces. Second, they usually neglect the high-order correlation contained in heterogeneous views. Third, many of them have three or even more hyper-parameters and may not be practical for some realistic applications. To tackle these issues, we present a new subspace clustering approach for incomplete multi-view data, namely Tensorized Incomplete Multiview Kernel subspace Clustering (TIMKC). By incorporating the kernel learning technique into an incomplete subspace clustering framework, our approach can robustly explore the latent subspace structure hidden in multiple views. Furthermore, the proposed approach imputes the incomplete kernel matrices and learns the low-rank tensor representations in a mutual enhancement manner. Notably, our approach can discover the underlying relationship among observed samples and missing samples, while capturing the high-order complementary information for robust subspace clustering. For solving the optimization model, we design a three-step algorithm to efficiently minimize the unified objective function, which only has one hyper-parameter to be adjusted. Experiments on various incomplete multi-view datasets demonstrate the superiority of our TIMKC approach. © 2023, The Authors. All rights reserved.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Near-Field Hierarchical Beam Management for RIS-Enabled Millimeter Wave Multi-Antenna Systems

TechRxiv

引用

TechRxiv 2022年

作者： Alexandropoulos, George C. Jamali, Vahid Schober, Robert Poor, H. Vincent Department of Informatics and Telecommunications National and Kapodistrian University of Athens Greece Department of Electrical and Computer Engineering Princeton University United States Germany

In this paper, we present a low overhead beam management approach for near-field millimeter-wave multi-antenna communication systems enabled by Reconfigurable Intelligent Surfaces (RISs). We devise a novel variable-width hierarchical phase-shift codebook suitable for both the near- and far-field of the RIS, and present a fast alignment algorithm for the RIS phase shifts and the transceiver beamformers. Indicative performance evaluation results are shown, verifying the effectiveness of the proposed approach in comparison with various benchmark schemes. © 2022, CC BY.

关键词： Millimeter waves

来源：评论

学校读者我要写书评

暂无评论

Sharding:A Scalability Solutions for Blockchain Networks

Sharding:A Scalability Solutions for Blockchain Networks

引用

Blockchain and Distributed Systems Security (ICBDS), IEEE International Conference on

作者： Dattatray Raghunath Kale Amolkumar N Jadhav Kruishna Prasad K Somanath Janardan Salunkhe Sumit Hirve Chandrashekhar Goswami Department of Computer Science & Engg MIT School of Computing MIT Art Design and Technology University Pune India Department of Computer Science & Engg Annasaheb Dange College of Engineering and Technology Ashta India Department of Computer Science & Engg Shrinivas University Mangalore India Department of Computer Science & Engg D Y Patil Agriculture and Technical University India Faculty of Computing and Informatics Sir Padampat Singhania University Udaipur Rajasthan

ISBN: (数字)9798350354348

ISBN: (纸本)9798350354355

Blockchain technology's decentralized and immutable data storage has changed a number of sectors. But typical blockchain networks scalability issues prevent them from being widely used for large-scale applications. By dividing the blockchain network into smaller, more controllable sections known as shards, sharding has become a viable remedy for scalability issues. This paper offers a thorough introduction to sharding as a blockchain network scalability solution. We explore the basic ideas of sharding as well as its advantages, drawbacks, and several sharding strategies. We also analyze experimental results and real-world implementations to assess how well sharding contributes to increased blockchain scalability. Lastly, we talk about possible developments in sharding approaches and future research paths to further improve the scalability of blockchain networks.

关键词： Industries Sharding Scalability Memory Throughput Blockchains Security

来源：评论

学校读者我要写书评

暂无评论

Hybrid Energy System Analysis for a Swimming Pool Complex using HOMER Pro

Hybrid Energy System Analysis for a Swimming Pool Complex us...

引用

International Universities Power engineering Conference (UPEC)

作者： Timea Farkas Paula Unguresan Mihaela Cretu Denisa Stet Levente Czumbil Andrei Ceclan Claudia Muresan Alexis Polycarpou Dan D. Micu Energy Transition Research Center Technical University of Cluj-Napoca Cluj-Napoca Romania Department of Electrical Engineering Computer Engineering and Informatics Frederick University Nicosia Cyprus

ISBN: (数字)9781665455053

ISBN: (纸本)9781665455060

The paper investigates hybrid energy systems that could be applied in case of a swimming pool complex in order to reduce exploitation costs and increase renewable energy usage. The presented approach starts with the assessment of the chosen building at the level of energy consumption, and utility costs, along with two proposed system configurations: one proposed by the RE-COGNITION innovation project, and one proposed by the authors based on the market solutions. The mentioned cases represent two hybrid energy systems using conventional sources, renewable energy, and highly efficient cogeneration units. The two proposed system configurations are simulated using the HOMER Pro program which allows a wide analysis along with techno-economic indicators of the different scenarios. Results and comparisons are presented to highlight the feasibility of the two proposed cases.

关键词： Renewable energy sources Analytical models Technological innovation Costs Technical management Sensitivity analysis Wind speed

来源：评论

学校读者我要写书评

暂无评论

Effective Subject Representation based on Multi-omics Disease Networks using Graph Embedding

Effective Subject Representation based on Multi-omics Diseas...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Hussein, Sundous Vu, Thao Langel, Leslie Bowler, Russell P. Kechris, Katerina J. Banaei-Kashani, Farnoush University of Colorado Denver Department of Computer Science and Engineering DenverCO United States University of Colorado Anschutz Medical Campus Department of Biostatistics and Informatics AuroraCO United States National Jewish Health Division of Pulmonary Critical Care and Sleep Medicine DenverCO United States University of Colorado Anschutz Medical Campus Division of Biomedical Informatics and Personalized Medicine AuroraCO United States

ISBN: (纸本)9781665468190

The study of complex behavior of biological systems has become increasingly dependent on evolutionary network modeling. In particular, multi-omics networks capture interactions between biomolecules such as proteins and metabolites, providing a basis for predicting relationships between such biomolecules and various phenotypic traits of complex diseases. In this paper, we introduce an integrative framework that given a multi-omics network representing a cohort of subjects, learns expressive representations for network nodes, and combines the learned nodes representations with the biological profiles of individual subjects for enriched representation of the subjects. With extensive empirical evaluation using real-world multi-omics networks, we show that our proposed framework significantly outperforms existing and baseline methods in terms of subject representation accuracy, particularly when the multi-omics network representing the cohort is sparse and structured and therefore, more informative. © 2022 IEEE.

关键词： Biomolecules

来源：评论

学校读者我要写书评

暂无评论

Correction: Harnessing AI for sustainable higher education: ethical considerations, operational efficiency, and future directions

引用

Discover Sustainability 2025年第1期6卷 1-2页

作者： Khan, Sunawar Mazhar, Tehseen Shahzad, Tariq Khan, Muhammad Amir Rehman, Ateeq Ur Saeed, Mamoon M. Hamam, Habib Department of Software Engineering Islamia University of Bahawalpur Bahawalnagar Pakistan School of Computer Science National College of Business Administration and Economics Lahore Pakistan Department of Computer Science School Education Department Government of Punjab Layyah Pakistan Department of Computer Engineering COMSATS University Islamabad Sahiwal Pakistan School of Computing Sciences College of Computing Informatics and Mathematics Universiti Teknologi MARA Shah Alam Malaysia School of Computing Gachon University Seongnam Republic of Korea Department of Communications and Electronics Engineering Faculty of Engineering University of Modern Sciences (UMS) Sana’a Yemen Faculty of Engineering Uni de Moncton Moncton Canada School of Electrical Engineering University of Johannesburg Johannesburg South Africa International Institute of Technology and Management (IITG) Av. Grandes Ecoles Libreville BP 1989 Gabon Bridges for Academic Excellence-Spectrum Tunis Tunisia

来源：评论

学校读者我要写书评

暂无评论

An Analysis of the Features Considerable for NFT Recommendations

arXiv

引用

arXiv 2022年

作者： Piyadigama, Dinuka Poravi, Guhanathan Computer Science and Engineering University of Westminster London United Kingdom Department of Computing Informatics Institute of Technology Colombo06 Sri Lanka

This research explores the methods that Nonfungible Token (NFT)s can be recommended to people who interact with NFT-marketplaces to explore NFTs of preference and similarity to what they have been searching for. While exploring past methods that can be adopted for recommendations, the use of NFT traits for recommendations has been explored. The outcome of the research highlights the necessity of using multiple Recommender Systems to present the user with the best possible NFTs when interacting with decentralized systems. © 2022, CC BY.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Learning for Resource Allocation in Wireless Networks of Dynamic Dimensionality

Multi-Task Learning for Resource Allocation in Wireless Netw...

引用

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

作者： Nikos A. Mitsiou Pavlos S. Bouzinis Panagiotis D. Diamantoulakis Panagiotis G. Sarigiannidis George K. Karagiannidis Department of Electrical and Computer Engineering Aristotle University of Thessaloniki Thessaloniki Greece MetaMind Innovations P.C Kozani Greece Department of Informatics and Telecommunications Engineering University of Western Macedonia Kozani Greece Artificial Intelligence & Cyber Systems Research Center Lebanese American University (LAU) Lebanon

ISBN: (数字)9798350362244

ISBN: (纸本)9798350362251

Deep neural networks (DNNs) have demonstrated their efficacy in delivering accurate solutions to a range of optimization problems. However, in the context of wireless communications, the size of these problems may vary across adjacent time slots, due to fast changes in the networks’ architecture, e.g., the number of users. It is essential to note that this time-varying dimensionality of optimization problems in wireless networks necessitates adjustments in the DNN architecture, resulting in different numbers of input and output nodes. To address this challenge, in our paper, optimization problems of varying size are treated as distinct tasks. To tackle these tasks, a multi-task learning (MTL) approach based on modular sharing is proposed. The multi-task approach consists of a DNN, which is used to extract the solutions for all the optimization problems, and a router which manages which nodes and layers of the input and output layer of the DNN to be used during the forward propagation of each task. Consequently, all tasks share common parameters of the DNN, while the DNN dynamically adjusts to the number of nodes of its output and input layers. Numerical results demonstrate the superiority of the suggested approach over zero-padding, which is the current solution for handling resource allocation problems of varying size.

关键词： Wireless networks Simulation Artificial neural networks Reinforcement learning Evolutionary computation Multitasking Dynamic scheduling Resource management Optimization Land mobile radio

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：