检索结果-内蒙古大学图书馆

An Analytical Review of Large Language Models Leveraging KDGI Fine-Tuning, Quantum Embedding’s, and Multimodal Architectures

引用

Computers, Materials and Continua 2025年第3期83卷 4031-4059页

作者： Uddagiri Sirisha Chanumolu Kiran Kumar Revathi Durgam Poluru Eswaraiah G Muni Nagamani Department of Computer Science and Engineering Prasad V Potluri Siddhartha Institute of Technology Vijayawada 520007 India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram 522302 India Department of Computer Science and Engineering (Data Science) AVN Institute of Engineering and Technology Hyderabad 501510 India Department of Computer Science and Engineering (Data Science) Vignan’s Institute of Management and Technology for Women Hyderabad 501301 India Department of Computer Science and Engineering Andhra Loyola Institute of Engineering and Technology Vijayawada 520008 India

A complete examination of Large Language Models’ strengths, problems, and applications is needed due to their rising use across disciplines. Current studies frequently focus on single-use situations and lack a comprehensive understanding of LLM architectural performance, strengths, and weaknesses. This gap precludes finding the appropriate models for task-specific applications and limits awareness of emerging LLM optimization and deployment strategies. In this research, 50 studies on 25+ LLMs, including GPT-3, GPT-4, Claude 3.5, DeepKet, and hybrid multimodal frameworks like ContextDET and GeoRSCLIP, are thoroughly reviewed. We propose LLM application taxonomy by grouping techniques by task focus—healthcare, chemistry, sentiment analysis, agent-based simulations, and multimodal integration. Advanced methods like parameter-efficient tuning (LoRA), quantum-enhanced embeddings (DeepKet), retrieval-augmented generation (RAG), and safety-focused models (GalaxyGPT) are evaluated for dataset requirements, computational efficiency, and performance measures. Frameworks for ethical issues, data limited hallucinations, and KDGI-enhanced fine-tuning like Woodpecker’s post-remedy corrections are highlighted. The investigation’s scope, mad, and methods are described, but the primary results are not. The work reveals that domain-specialized fine-tuned LLMs employing RAG and quantum-enhanced embeddings perform better for context-heavy applications. In medical text normalization, ChatGPT-4 outperforms previous models, while two multimodal frameworks, GeoRSCLIP, increase remote sensing. Parameter-efficient tuning technologies like LoRA have minimal computing cost and similar performance, demonstrating the necessity for adaptive models in multiple domains. To discover the optimum domain-specific models, explain domain-specific fine-tuning, and present quantum and multimodal LLMs to address scalability and cross-domain issues. The framework helps academics and practitioners identify, a

关键词： Large language models quantum embeddings fine-tuning techniques multimodal architectures ethical AI scenarios

来源：评论

学校读者我要写书评

暂无评论

Online Non-convex Optimization with Long-term Non-convex Constraints

arXiv

引用

arXiv 2023年

作者： Pan, Shijie Huang, Wenjie The Department of Industrial Manufacturing and Systems Engineering The University of Hong Kong Pokfulam Road 999077 Hong Kong The Musketeers Foundation Institute of Data Science The Department of Industrial Manufacturing and Systems Engineering The University of Hong Kong Pokfulam Road 999077 Hong Kong

A novel Follow-the-Perturbed-Leader type algorithm is proposed and analyzed for solving general long-term constrained optimization problems in online manner, where the objective and constraints are arbitrarily generated and not necessarily convex. In each period, random linear perturbation and strongly concave perturbation are incorporated in primal and dual directions, respectively, to the offline oracle, and a global minimax point is searched as the solution. Based on a proposed expected static cumulative regret, we derive the first sublinear O(T8/9) regret complexity for this class of problems. The proposed algorithm is applied to tackle a long-term (extreme value) constrained river pollutant source identification problem, validate the theoretical results and exhibit superior performance compared to existing methods. Copyright © 2023, The Authors. All rights reserved.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Hyper-parameter Tuning of CNN Using Improved Elephant Herding Optimisation for Detection of Skin Cancer 3rd

Hyper-parameter Tuning of CNN Using Improved Elephant Herdin...

引用

3rd International Conference on Computing and Communication Networks, ICCCN 2023

作者： Uma, N. Siva Shankar, G. Kavin, Balasubramanian Prabhu Dhanaraj, Rajesh Kumar New Horizon College of Engineering Ring Road Bellandur Post Bengaluru560103 India Department of Computational Intelligence College of Engineering and Technology SRM Institute of Science and Technology Tamil Nadu Kattankulathur Chengalpattu District 603203 India Department of Data Science and Business Systems College of Engineering and Technology SRM Institute of Science and Technology Tamil Nadu Kattankulathur Chengalpattu District 603203 India Pune India

ISBN: (纸本)9789819726707

Skin cancer is one of the top three tumours caused by DNA damage and has a dire prognosis. As a result of this injured DNA, cells start to grow out of control and are currently doubling in size. This study recommends using a hyper-parameter optimised convolution neural net to identify the kind of skin cancer. In this method, the improved elephant herding optimisation (IEHO) algorithm was used to optimise CNN's hyper-parameters. A nonlinear denoising method known as a bilateral filter can lessen noise while maintaining edges. In this study, we frequently use the bilateral filter to modify the texture and noise in skin cancer photo images. An improved elephant herding optimisation has been proposed (IEHO). When the separation operator was changed to the sine–cosine, opposition-based learning was incorporated. The proposed model, which is around 4 and 1% higher than other based models, may produce testing accuracy up to 99.33%, according to simulation results. The experimental findings unequivocally demonstrate that the recommended model outperforms other published models. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Development of A Multipurpose Healthcare Mobile App Using JAVA

Development of A Multipurpose Healthcare Mobile App Using JA...

引用

2023 International Conference on System, Computation, Automation and Networking, ICSCAN 2023

作者： Pongiannan, R.K. Bushra, S. Nikkath Lathif, Syed Ismail Abdul Pradhan, Saahil Brindha, R. Franklin, J. Srm Institute of Science and Technology Department of Computing Technologies Kattankulathur India Data Science and Business Systems School of Computing Srm Institute of Science and Technology Kattankulathur India Srm Institute of Science and Technology Department of Computational Intelligence Kattankulathur India Srm Institute of Science and Technology Department of Electrical and Electronics Engineering Kattankulathur India

ISBN: (纸本)9798350315127

Health care has become an essential part of present human life. The preliminary requirement to select the basic information about medical facilities through a mobile app will serve the needs of the common people. This paper presents an integrated mobile app developed using JAVA. This app is user-friendly in terms of login, registration, and bookings. This caters to the facility to select information regarding doctors, medical lab facilities, pharmacy details, and health articles. This app will provide communication. © 2023 IEEE.

关键词： Medicine

来源：评论

学校读者我要写书评

暂无评论

An Advanced Intrusion Detection Solution for Networks based on Honeypot Servers 6

An Advanced Intrusion Detection Solution for Networks based ...

引用

6th International Conference on Inventive Computation Technologies, ICICT 2023

作者： Veena, K. Meena, K. Mercy Theresa, M. Hemalatha, C. Rajalakshmi, D. Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai600119 India Department of Computer Science and Engineering GITAM School of Technology GITAM Bangalore561203 India Department of Data Science and Business Systems School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai603203 India Computer Science and Engineering Sri Sairam Institute of Technology Chennai India

ISBN: (纸本)9798350398496

Network security is a critical issue in modern technology. Honey pot-based intrusion detection methods provide an additional layer of security and enhance network performance by analyzing hacker behaviour and detecting unauthorized clients. IP validation and vulnerability detection are used to ensure authorized customer access and prevent illegal client activity. A voice recognition system detects malicious attacks in stage two. If authorized, packets are passed to the server, otherwise, they are delivered to honeypot servers. Using Naive Bayes algorithm, the detection rate is 85.06% with an accuracy of 98.18% and a false alarm rate of 0.15 seconds. © 2023 IEEE.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Despise Detection through LOA Using Fs with 1D-CNN in Public Forum 5

Despise Detection through LOA Using Fs with 1D-CNN in Public...

引用

5th IEEE International Conference on Computing, Power, and Communication Technologies, IC2PCT 2024

作者： Revathi, V. Gunapriya, B. Kavin, Balasubramanian Prabhu Gangadevi, E. Balusamy, Balamurugan Gite, Shilpa New Horizon College of Engineering Ring Road Bellandur Post Bengaluru India Srm Institute of Science and Technology Department of Data Science and Business Systems Chengalpattu Kattankulathur India Loyola College Department of Computer Science Chennai India Shiv Nadar University Delhi India Symbiosis Institute of Technology Pune India

ISBN: (纸本)9798350383522

Hate speech is any speech intended to incite hatred in others. Hate speech is escalating quickly and unpredictably in this digital age where everyone is connected via social media. Due to a lack of understanding of the distinction among hate speech numerous people do not realise they are engaging in hate speech when they criticise somewhat on social media. Victims experience a sense of social alienation as a result, and those who propagate it frequently end up in legal trouble. To combat people's ignorance, it is crucial to detect any instances of hate speech in the sentences. A machine learning technique is frequently used to help classify each sentence in order to detect such sentences. Additionally, it is a difficult duty because to the enormous volume of rising hate speech on social media from diverse sources. This work offers a Lion optimisation Algorithm (LOA) with Processing Based Hate Speech Detection (HSD) and Classification model in response to this objective. The main goal of the proposed LOA-based HSD model is to recognise and categorise instances of hate speech on social media platforms. The described model uses data preprocessing at numerous phases, such as tokenization, vectorization, etc., to achieve this. In addition, the organisation of social media text into three categories-neutral, offensive, and hate language-is done using a convolutional neural network (ID-CNN) model. The proposed model is experimentally validated, and the outcomes are assessed from several angles. The results of the experiment showed that the LOA-ID-CNN model achieved favourably compared to current state-of-the-art techniques © 2024 IEEE.

关键词： Convolutional Neural Network Hate Speech Finding Lion Optimization Algorithm social media Twitter. Natural Language Processing

来源：评论

学校读者我要写书评

暂无评论

Position: scarce resource allocations that rely on machine learning should be randomized 24

Position: scarce resource allocations that rely on machine l...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Shomik Jain Kathleen Creel Ashia Wilson Institute for Data Systems and Society MIT Department of Philosophy & Religion and Khoury College of Computer Sciences Northeastern University Institute for Data Systems and Society and Department of Electrical Engineering and Computer Science MIT

Contrary to traditional deterministic notions of algorithmic fairness, this paper argues that fairly allocating scarce resources using machine learning often requires randomness. We address why, when, and how to randomize by proposing stochastic procedures that more adequately account for all of the claims that individuals have to allocations of social goods or opportunities.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Load and Routes in Supply Chain Management with Travelling Salesman Problem and CapsNet Attention 3

Optimizing Load and Routes in Supply Chain Management with T...

引用

3rd International Conference on Integrated Circuits and Communication systems, ICICACS 2025

作者： Rajasekar, P. Begam, M. Shabeena Chattopadhyay, Shouvik Basha, S. Chand Chandre, Shanker Dwivedi, Vijay Kumar Srm Institute of Science and Technology Department of Data Science and Business Systems Kattankulathur India Sns College of Technology Department of Mba Coimbatore India Institute of Engineering & Management Department of Mba Kolkata India University of Engineering & Management Kolkata India St Ann's College of Engineering and Technology Department of Business Administration Chirala India School of Computer Science & Artificial Intelligence Sr University Department of Computer Science & Artificial Intelligence Warangal India Vishwavidyalaya Engineering College Department of Mathematics Ambikapur India

ISBN: (纸本)9798331508456

Ever since the industrial revolution, industries have partnered to solve marketing problems and create new channels. Because they don't understand their clients' needs, sales centers often pursue conflicting aims. Lack of information has extended delivery times. Due to supply time, quality, or bulk freight depreciation agreements, facility expenses rise to satisfy demand, resulting in excess inventory. Supply chain management, particularly regional inventory management, has considerably impacted beer manufacturing profitability. The recommended method involves training the model and extracting features. Dynamic and PCA feature extraction are different. While PCA's main purpose is to dissociate correlated from uncorrelated data, dynamic PCA considers auto- and cross-correlation in time-series observations. A revised CapsNet approach trained the models. The Improved CapsNet model outperformed the Attention and baseline models in every test. The model solved brewing supply chain issues with 94.68% accuracy. The most successful advanced machine learning approaches for improving SCM include augmented CapsNet, PCA, and dynamic PCA, according to research. These techniques boost industrial profitability when applied to major issues like excess stock and supply unpredictability. The suggested strategy is more reliable and superior than present solutions, according to results. © 2025 IEEE.

关键词： Supply chain management

来源：评论

学校读者我要写书评

暂无评论

Dynamic Topology Management In Ad-Hoc Networks For Improved Performance 2

Dynamic Topology Management In Ad-Hoc Networks For Improved ...

引用

2nd IEEE International Conference on Advances in Information Technology, ICAIT 2024

作者： Anakath, A.S. Kannadasan, R. Margarat, G. Simi Pandian, A. Pasumpon Antony Sibiya Varghese, V. Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tamil Nadu Chennai India Department of Software Systems School of Computer Science and Engineering Vellore Institute of Technology VIT University Tamil Nadu Vellore India Department of Computer Science and Engineering New Prince Shri Bhavani College of Engineering and Technology Tamil Nadu Chennai India Department of Computer Science and Engineering CARE College of Engineering Trichy India Department of Artificial Intelligence and Data Science GRT Institute of Engineering and Technology Tamil Nadu Tiruttani India

ISBN: (纸本)9798350383867

Optimization in ad hoc networks is a highly specialized task because the structure of the network is loosely formed, and each node is independently responsible for its operation. In order to obtain a better result from this solely formed networks this report offer extensive Dynamic Topology Management (DTM) method. Dynamic routing protocols lager lines, dividend balancing, Life-saving administration, guarantee of reliability, cross-layer design, security aspects and the on-time monitoring and discovery of topology changes are the main objectives and steps to be accomplished. Real-time Location monitoring and forest structure discovery are made possible by employing adaptive routing protocols, namely Dynamic Source Routing(DSR) and Ad-Hoc On-Demand Distance Vector(AODV). Also, the algorithms are capable of making advance adjustments to routing routes and resource allocations drawn from the mobility patterns forecasting which is vital. Load balancing methods, which distribute traffic in a network evenly, enable the system to run efficiently and avoid problems caused by overloading. Algorithms that are energy-efficient and make the best use of energy resources having their topology's optimized are introduced to the streamlined energy distribution systems. These algorithms are focused on the node's energy level. Through fault tolerance techniques which recognize and handle failures in nodes and partitioned networks, connectedness activity is a continuous one. Through QoS perception, the network topology might automatically be tailored to conform to the given quality criteria, thus not restricting the choice for differing applications. The proposed could have defenses against adversary nodes and their potential DDoS attacks which will enhance network protection. Yet another advantage of an architectural style that promotes cross-layer collaboration is enhanced topology management which is the goal around which the functions of various levels of the protocol stack revolve

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

PLAME: Piecewise-Linear Approximate Measure for Additive Kernel SVM (Extended abstract)

PLAME: Piecewise-Linear Approximate Measure for Additive Ker...

引用

International Conference on data engineering

作者： Tsz Nam Chan Zhe Li Leong Hou U Reynold Cheng Department of Computer Science and Software Engineering Shenzhen University Alibaba Cloud Department of Computer and Information Science State Key Laboratory of Internet of Things for Smart City University of Macau Department of Computer Science and Musketeers Foundation Institute of Data Science The University of Hong Kong

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Additive Kernel SVM has been extensively used in many applications, including human activity detection and pedestrian detection. Since training an additive kernel SVM model is very time-consuming, which is not scalable to largescale datasets, many efficient solutions have been developed in the past few years. However, most of the existing methods normally fail to achieve one of these three important conditions which are (1) low classification error, (2) low memory space, and (3) low training time. In order to simultaneously fulfill these three conditions, we develop the new piecewise-linear approximate measure (PLAME) for training additive kernel SVM models. Experimental results verify that this approach can achieve the best trade-off between accuracy, memory space, and training time compared with different types of state-of-the-art methods.

关键词： Training Support vector machines Additives Pedestrians Accuracy data engineering Kernel

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：