检索结果-内蒙古大学图书馆

Knowledge-Empowered,Collaborative,and Co-Evolving AI Models:The Post-LLM Roadmap

工程（英文） 2025年第1期44卷 87-100页

作者： Fei Wu Tao Shen Thomas Bäck Jingyuan Chen Gang Huang Yaochu Jin Kun Kuang Mengze Li Cewu Lu Jiaxu Miao Yongwei Wang Ying Wei Fan Wu Junchi Yan Hongxia Yang Yi Yang Shengyu Zhang Zhou Zhao Yueting Zhuang Yunhe Pan College of Computer Science and Technology Zhejiang UniversityHangzhou 310027China Leiden Institute of Advanced Computer Science Leiden UniversityLeiden 2333 CCNetherlands College of Electrical Engineering Zhejiang UniversityHangzhou 310027China School of Engineering Westlake UniversityHangzhou 310024China Department of Computer Science and Engineering Hong Kong University of Science and TechnologyHong Kong 999077China Department of Computer Science and Engineering Shanghai Jiao Tong UniversityShanghai 200240China School of Cyber Science and Technology Sun Yat-Sen UniversityShenzhen 518107China Department of Computing The Hong Kong Polytechnic UniversityHong Kong 999077China

Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple *** these achieve-ments,LLMs have inherent limitations including outdated information,hallucinations,inefficiency,lack of interpretability,and challenges in domain-specific *** address these issues,this survey explores three promising directions in the post-LLM era:knowledge empowerment,model collaboration,and model ***,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 ***,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 ***,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 *** 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 *** conclude by outlining future pathways for further advancement and applications.

关键词： Artificial intelligence Large language models Knowledge empowerment Model collaboration Model co-evolution

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Survey on Methods Used for Automatic Assessment of Subjective Answers

A Comprehensive Survey on Methods Used for Automatic Assessm...

引用

2024 IEEE International Students' Conference on Electrical, Electronics and computer Science, SCEECS 2024

作者： Joshi, Sudarshan Bachkar, Akshay Awaje, Omkar Bhoir, Rhutuj Savitribai Phule Pune University Pune Institute of Computer Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350348460

The paper explores various techniques for identifying multilingual text similarities and basic image similarities, covering a wide array of methods and opportunities for comparing and evaluating textual and diagrammatic content. By delving into these categories and studying the pros and cons of various methods, the survey equips researchers, practitioners, and enthusiasts in the field of NLP with a deeper understanding of text similarity *** paper delves into tokenization, vectorization, and similarity metrics, among other approaches. It also discusses modelling techniques, such as LSA and LDA, text alignment algorithms, cosine similarity, BERT, and optimization techniques. The survey categorizes these techniques into corpus-based, knowledge-based, semantic, term-based, and character-based similarities. The survey aims to provide a comprehensive overview of the methods available for checking multilingual text similarity identification and its significance in contemporary text analysis. Index Terms - Similarity Techniques, Machine learning, Natural Language Processing. © 2024 IEEE.

关键词： Image Similarity Multilingualism Natural Language Processing Semantic Analysis Similarity Techniques

来源：评论

学校读者我要写书评

暂无评论

Localize the Copy-Move Forged Region of an Image Using Improved SIFT

引用

SN computer Science 2024年第1期5卷 1-15页

作者： Dhanaraj, Rachel Selva Sridevi, M. Department of Computer Science and Engineering National Institute of Technology Tamil Nadu Tiruchirappalli India

With the advancement in technology, digital images have become a very popular source of information. But this source of information is also being used to mislead people and society by altering the original image. One of the ways through which digital images are tampered with is the copy-move forgery method, wherein a part of an image is copied and pasted into another area inside the same image. To avoid and prevent any kind of anomalies, this paper focuses on ways to detect this copy and move forgery in an image using the keypoint-based approach in an efficient manner. The methodology involves detecting key points using Improved Scale Invariant Feature Transform (SIFT) based on which features are identified. Within the isolated descriptors, the deployed K-Nearest Neighbor (K-NN) classifier determines the closest match. Now in order to remove the outliers present in the features, Hierarchical Clustering is used to cluster similar descriptors. Experimental result has proved this methodology to be effective in detecting the forgery in an image for different kinds of related transformations. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Copy-move forgery Hierarchical clustering K-NN classifier Scale invariant feature transform

来源：评论

学校读者我要写书评

暂无评论

NLP based Text Summarization of Fintech RFPs 2

NLP based Text Summarization of Fintech RFPs

引用

2nd International Conference on Sustainable Computing and Data Communication Systems, ICSCDS 2023

作者： Patil, Krutika Badamikar, Medha Sonawane, Sheetal Pune Institute of Computer Technology Department of Computer Engineering Pune India

ISBN: (纸本)9781665491990

In today's day and age, where huge quantities of textual data are generated every second, it has become difficult to keep ourselves abreast with new information. Documents in the financial sector tell a quantitative story. The qualitative or language content that accompanies financial statements is an essential component of the information set used by financial market participants for monitoring and stewardship. This necessitates the development of efficient technological methods for leveraging the existence of these massive amounts of textual data. Reading financial documents such as annual reports is extremely time-consuming and thus corporations have to allocate precious human capital to understand, analyze and derive insight from these reports. As a result, automatic summarization techniques can make this task simpler by enabling access to a smaller but informative portion of a given document. In this work, a system utilizing NLP techniques for summarizing financial documents based on queries provided by the user, is presented. This system overcomes the challenges posed by existing section-based summarization models. As machine learning methods have proven to be effective on downstream tasks such as text summarization, the intent of this work is to take advantage of these methods for effective extractive summarization and generate human-like summaries. The proposed model categorizes phrases according to their relevance by leveraging the strength of unsupervised clustering approaches, and it gets a ROUGE-1 score of 46%. © 2023 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Effective & Smart Urban Mobility System at Low Cost 2

Effective & Smart Urban Mobility System at Low Cost

引用

2nd International Conference Computational and Characterization Techniques in engineering and Sciences, IC3TES 2024

作者： Jyothula, Hari Rao, Y Laxmana Prasad, K. Vara Usha, Chiriki Manikyamba, I Lakshmi Aditya University Department of Computer Science and Engineering Surampalem India Vignan's Institute of Engineering for Women Department of It Visakhapatnam India Avanthi Institute of Engineering & Technology I Department of Computer Science and Engineering India Dadi Institute Of Engineering And Technology Department of Computer Science and Engineering India University College of Engineering Science & Technology Hyderabad Jawaharlal Nehru Technological University Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798350364699

There is no established metric for measuring urban smartness at this time. Various writers use various metrics to assess smart city transportation systems. The assessment indicators for the transport system that were offered in the scholarly works were reviewed, and the most relevant and impactful indicators were chosen. A hierarchical assessment model for smart city transport systems is developed in this paper. There are five categories, or "factors,"that the indications fall into. Several indicators are assigned to each of the categories. This study used the hybrid MCDM (multi-criteria decision-making) method to evaluate urban transport systems and identify which of the given characteristics were most important. Top Priority Sorting based on Ideal Solution Similarity (TOPSIS), COPRAS, and Simple Additive Weighting (SAW) were the multi-criteria assessment methods in use. The importance of the components and indicators was determined using an analytical hierarchy process (AHP), rankings (prioritization), and direct evaluation (experts evaluate each criterion as a percentage;total assessments of all criteria equal 100%). There were two phases to the evaluation and comparison of mobility systems: first, smart mobility systems were evaluated comprehensively according to the integrated importance of the indicators;and second, multi-criteria evaluations were conducted according to the indicators of every factor separately. Based on the amount of intelligence, a top city is determined and rated. Building a hierarchical model for evaluating smart transportation systems is the focus of this paper. © 2024 IEEE.

关键词： Evaluation smart transportation MCDM Smart City Smart Mobility System Urban smartness

来源：评论

学校读者我要写书评

暂无评论

Real-Time Electrochemical Model-Based BMS Control for Mitigating Li-Plating and Extending Battery Life

引用

IEEE Transactions on Transportation Electrification 2025年第3期11卷 7403-7419页

作者： Qasem, Mohammad Haddadin, Mariana Yassin, Yazan Ratrout, Sadam Chen, Chengxiu Stoyanov, Stoyan Al-Hallaj, Said Krishnamurthy, Mahesh Southern Methodist University Department of Electrical and Computer Engineering DallasTX75205 United States Illinois Institute of Technology Department of Electrical and Computer Engineering ChicagoIL60616 United States Beam Global BroadviewIL60155 United States

It is well-known that lithium plating significantly reduces the capacity of Li-ion batteries, particularly at elevated charging rates, high state of charge (SoC), and low temperatures. This study presents a simplified electrochemical battery model incorporating side reactions for real-time (RT) applications. The proposed model is implemented on a battery management system (BMS) using the TI TMS320F28335 DSP, the model’s internal and external states are validated through extensive experimental tests against the high-fidelity P2D model by Gamma technology. The results demonstrated the model’s ability to accurately depict the behavior of a commercially available type 21 700 cylindrical cell while minimizing execution power consumption. The lithium plating/stripping model is examined and compared with the actual voltage plateau characteristics observed in practical testing. Finally, various controllers are designed and implemented into the BMS to optimize the charging current and eliminate plating without significantly increasing charging duration. The battery demonstrated an increase in capacity of 8.3% when using the proposed charging control algorithm compared with the common constant current (CC)–constant voltage (CV) technique, with a charging time increase of less than 30 min, resulting in a 17% improvement in SoH. This demonstrates the effectiveness of enhancing battery performance and longevity with minimal additional charging time. © 2015 IEEE.

关键词： Lithium-ion batteries

来源：评论

学校读者我要写书评

暂无评论

An Integrated Approach of Efficient Edge Task Offloading Using Deep RL, Attention and MDS Techniques

引用

SN computer Science 2024年第6期5卷 1-15页

作者： Priyadarshni Kumar, Praveen Kadavala, Dhruvan Tripathi, Shivani Misra, Rajiv Department of Computer Science and Engineering Indian Institute of Technology Patna Bihar Patna India

In Distributed Computation Optimization (DCO) networks, where clients distribute computational jobs among heterogeneous helpers with different capacities and pricing models, efficient task offloading and cost reduction continue to be major issues. In order to tackle this problem, a unique method utilizing Deep Reinforcement Learning (DRL) is presented in this study. The DRL algorithm allows clients to independently identify the best method of action without requiring prior knowledge of network dynamics since it adjusts to the dynamic and stochastic nature of DCO environments. In distributed computing environments, the combination of Maximum Distance Separable (MDS) and DRL approaches provides a stable framework for task offloading. The suggested method finds the best policy for helper selection and task offloading by combining reward estimation, action selection, environment modeling, and iterative learning approaches. In particular, energy consumption, density, and total rewards show that the DRL algorithm performs better in experimental evaluations than traditional techniques like Q-learning and random selection. The algorithm’s performance is further improved by the integration of DQN with the attention mechanism, highlighting its potential to completely transform the efficiency of DCO networks. This work highlights how DRL approaches have a significant effect on network operations optimization and offers insightful information about future developments in distributed computation paradigms. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Deep reinforcement learning Distributed computation optimization Edge computing MDS techniques Task offloading

来源：评论

学校读者我要写书评

暂无评论

DL Based Speech to Text Converter for Audio Visual Applications 3

DL Based Speech to Text Converter for Audio Visual Applicati...

引用

3rd IEEE International Conference on ICT in Business Industry and Government, ICTBIG 2023

作者： Kulkarni, Rekha Takalikar, Mukta Awatade, Nikhil Bhalgat, Ameya Pune Institute of Computer Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350343274

In the digital era, Speech recognition and Text conversion has become a prominent research field with numerous applications. The accurate recognition of speech and quick conversion possess number of challenges like accent, mixed words, variation in pronunciation etc. The proposed system leverages the power of machine learning algorithms to accurately recognize speech and convert it into text in real time. The system is trained using the ML modules like speech recognition, moviepy etc. Overall, our work demonstrates the potential of machine learning-based approaches for speech recognition and text conversion and highlights the need for further research in this area to address the challenges and limitations of existing systems. © 2023 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

ICE-VDOP: an integrated clustering and ensemble machine learning methods for an enhanced vector-borne disease outbreak prediction using climatic variables

引用

International Journal of Information technology (Singapore) 2024年第4期16卷 2077-2088页

作者： Jayabalan, Divya Elango, Sivasankar Department of Computer Science and Engineering National Institute of Technology Tamilnadu Tiruchirappalli 620015 India

Dengue is prone to cause frequent disease outbreaks. An early outbreak prediction based on the climatic variables is essential to control the disease’s spread and avoid possible outbreaks. Existing non-seasonal outbreak classification uses a partition-based clustering method for outlier removal;however, a standard outlier detection with improved prediction performance is needed to detect outbreaks. We develop ICE-VDOP model, an Integrated Clustering and Ensemble machine-learning model for Vector-borne Disease Outbreak Prediction to classify outbreaks based on non-seasonal climatic characteristics for Bangkok and Bangladesh datasets. The work begins with identifying the clustering approach that generates a refined dataset ensuring the data points’ inherent structure is preserved and does not alter the association of climatic variables and Dengue Incidence. The partitioning-based clustering methods modify the direction of association between climatic variables and Dengue Incidence after outlier removal as it removes the essential data points. The comparative results shows that BIRCH (balanced iterative reducing and clustering hierarchy), a hierarchical clustering technique for outlier removal captures finer non-seasonal climate variables (temperature, rainfall and relative humidity) and gradient boosting (GB) achieves a prediction accuracy of 97% for Bangkok and 98% for the Bangladesh datasets. © The Author(s), under exclusive licence to Bharati Vidyapeeth's institute of computer Applications and Management 2024.

关键词： Binary classification BIRCH Dengue Gradient boosting Outlier detection Vector-borne disease outbreak prediction

来源：评论

学校读者我要写书评

暂无评论

Multimodal Deep Learning Framework for Non-Intrusive Autism Spectrum Disorder Detection 3

Multimodal Deep Learning Framework for Non-Intrusive Autism ...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Kavitha, G. Rajganesh, N. Rajalakshmi Institute of Technology Department of Computer Science and Engineering Chennai India Sri Venkateswara College of Engineering Department of Computer Science and Engineering Sriperumbudur Chennai India

ISBN: (纸本)9798331509675

The non-intrusive detection of Autism Spectrum Disorder (ASD) marks a significant advancement in early diagnosis and intervention. This approach allows users to upload videos to a web interface, where visual and auditory data are processed separately. Visual data is analyzed using a customized VGG19 Convolutional Neural Network (CNN) to identify facial expressions and movements indicative of autism, while auditory data is evaluated by a Recurrent Neural Network (RNN) to detect auditory cues like tone and rhythm. A voting mechanism across 10 frames determines the visual classification, labelling the subject autistic if most frames indicate autistic traits. Audio analysis provides an independent prediction. The subject is classified as autistic if either analysis indicates autistic features. By leveraging deep neural networks to process multimodal data, this method offers an accessible, early ASD detection tool for caregivers and professionals, promoting timely interventions and improved outcomes. © 2025 IEEE.

关键词： Convolutional neural networks

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：