检索结果-内蒙古大学图书馆

Multimodal Large Models Are Effective Action Anticipators

IEEE Transactions on Multimedia 2025年 27卷 2949-2960页

作者： Wang, Binglu Tian, Yao Wang, Shunzhou Yang, Le College of Information and Control Engineering Xi'an University of Architecture and Technology Xi'an710055 China Northwestern Polytechnical University School of Astronautics Xi'an710072 China Peking University School of Electronic and Computer Engineering Shenzhen Graduate School Shenzhen518055 China Hefei Comprehensive National Science Center Institute of Artificial Intelligence China University of Science and Technology of China School of Information Science and Technology Hefei230022 China

The task of long-term action anticipation demands solutions that can effectively model temporal dynamics over extended periods while deeply understanding the inherent semantics of actions. Traditional approaches, which primarily rely on recurrent units or Transformer layers to capture long-term dependencies, often fall short in addressing these challenges. Large Language Models (LLMs), with their robust sequential modeling capabilities and extensive commonsense knowledge, present new opportunities for long-term action anticipation. In this work, we introduce the ActionLLM framework, a novel approach that treats video sequences as successive tokens, leveraging LLMs to anticipate future actions. Our baseline model simplifies the LLM architecture by setting future tokens, incorporating an action tuning module, and reducing the textual decoder layer to a linear layer, enabling straightforward action prediction without the need for complex instructions or redundant descriptions. To further harness the commonsense reasoning of LLMs, we predict action categories for observed frames and use sequential textual clues to guide semantic understanding. In addition, we introduce a Cross-Modality Interaction Block, designed to explore the specificity within each modality and capture interactions between vision and textual modalities, thereby enhancing multimodal tuning. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed ActionLLM framework, encouraging a promising direction to explore LLMs in the context of action anticipation. © 1999-2012 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Development of Dementia Care Training System Using AR and Large Language Model

Development of Dementia Care Training System Using AR and La...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Miyawaki, Tomoya Nishiura, Yuuki Fukuda, Ryouta Nakashima, Kazuto Kurazume, Ryo Graduate School of Information Science and Electrical Engineering Kyushu University Japan Kyushu University Faculty of Engineering Department of Electrical Engineering and Computer Science Japan Kyushu University Faculty of Information Science and Electrical Engineering Japan

ISBN: (纸本)9781665410205

We have developed HEARTS, a dementia care training system using augmented reality based on Humanitude. Humanitude is a multimodal comprehensive care technique for dementia, and has attracted attention as a method to reduce the burden on both caregivers and patients. However, the HEARTS developed so far could not evaluate 'speaking' skills based on the content of conversations among 'seeing,' 'touching,' and 'speaking,' all of which are fundamental skills in Humanitude. Therefore, we attempted a new quantitative evaluation of trainees' 'speaking' skills by estimating the emotional value of conversational content using GPT-4 named HEARTS 5. A survey of caregivers was conducted using the developed system and was well received by the participants. We also developed a HEARTS 5 conversational version based on HEARTS 5, with the addition of GPT-4 conversation generation and Azure Text-to-Speech. © 2024 IEEE.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Mobile Application to Help Abandoned Pets 3

Mobile Application to Help Abandoned Pets

引用

3rd IEEE International Conference on Blockchain and Distributed Systems Security, ICBDS 2024

作者： Rahajeng, Salicca Dewi Nurfitriyani, Siti Julianingsih Akbar, Zakin Izzani Leonardo, Calvin Warnars, Harco Leslie Hendric Spits Warnars, Laurens Spits Bina Nusantara University School of Information Systems Information Systems Department Jakarta11480 Indonesia Bina Nusantara University Binus Graduate Program - Computer Science Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798350354348

As social beings, humans like to interact with each other, including other creatures of God, such as animals, and keep them as pets. Many pets, such as dogs, cats, birds, fish, rabbits, and so on, include unusual pets such as reptiles, pigs, insects, etc. As a result of the lack of human awareness of pets and the nature of keeping animals, which is only for fun, it is not soulful to keep animals. So when people are bored, they abandon their pets, which causes new problems in human life today. In this case, we created a mobile application that can be shared with pet lovers where they can share information about abandoned pets and the possibility of owning pets from these abandoned pets. Mobile applications were designed using use case diagrams to show business processes, class diagrams to show database models, and user interfaces to present a similar interface to users. This mobile application will become a platform that can gather all people who care about pets and make pets a part of humans. © 2024 IEEE.

关键词： Mobile applications

来源：评论

学校读者我要写书评

暂无评论

VLC-UNIT: Unsupervised Image-to-Image Translation with Vision-Language Classification 17

VLC-UNIT: Unsupervised Image-to-Image Translation with Visio...

引用

17th IEEE International Symposium on Embedded Multicore/Many-core Systems-on-Chip, MCSoC 2024

作者： Liang, Yuying Huang, Huakun Wang, Bin Zhao, Lingjun Xu, Jiantao Zhang, Chen Guangzhou University School of Computer Science and Cyber Engineering Guangzhou China Guangdong Polytechnic Normal University Electronics and Information Guangzhou China University of Aizu Graduate School of Computer Science and Engineering Japan

ISBN: (纸本)9798331530471

In recent years, unsupervised image-to-image translation technology has garnered widespread attention for its creative expression, accompanied by enhanced data augmentation capabilities. However, current methods are heavily influenced by dataset distributions, resulting in reduced performance on data-scarce samples and inconsistencies in fine-grained categories and target styles, leading to the generation of pseudo categories. To tackle these challenges, we introduce a novel framework called Vision-Language Classification (VLC-UNIT) for UNsupervised Image-to-Image Translation. By leveraging prompts from large vision-language models, VLC-UNIT enhances the semantic understanding of images and improves the consistency and accuracy of image translations. Experimental results demonstrate our VLC-UNIT outperforms existing state-of-the-art techniques. © 2024 IEEE.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

Generating High-Quality Distributed Representations of Words Using Wiktionary and Sentence-BERT 16

Generating High-Quality Distributed Representations of Words...

引用

16th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2024

作者： Nishiura, Ryota Tsuchiya, Seiji Watabe, Hirokazu Graduate School of Science and Engineering Doshisha University Dept. of Information and Computer Science Kyotanabe Kyoto Japan Doshisha University Faculty of Science and Engineering Dept. of Intelligent Information Engineering and Science Kyotanabe Kyoto Japan

ISBN: (纸本)9798350377903

We propose an improved method to generate high-quality distributed representations of words. In our previous paper, we introduced a novel method to obtain distributed representations of words with BERT using Wiktionary as semantic sources of words. The previous method outperformed a Wrd2Vec model in a word similarity task, while there still had to be some room to improve in term of semantic accuracy. In this paper, our new method introduces one of Sentence-BERT models to text-embed the semantic sources instead of the original BERT model. We demonstrate that our proposed method outperforms both our previous method and the Word2Vec model in the tasks. © 2024 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Prompting Large Language Models with Knowledge-Injection for Knowledge-Based Visual Question Answering

引用

Big Data Mining and Analytics 2024年第3期7卷 843-857页

作者： Zhongjian Hu Peng Yang Fengyuan Liu Yuan Meng Xingyu Liu School of Computer Science and Engineering Southeast University the Key Laboratory of Computer Network and Information Integration(Southeast University) Ministry of Education of the People’s Republic of ChinaNanjing 211189China Southeast University-Monash University Joint Graduate School(Suzhou) Southeast UniversitySuzhou 215125China

Previous works employ the Large Language Model(LLM)like GPT-3 for knowledge-based Visual Question Answering(VQA).We argue that the inferential capacity of LLM can be enhanced through knowledge *** methods that utilize knowledge graphs to enhance LLM have been explored in various tasks,they may have some limitations,such as the possibility of not being able to retrieve the required *** this paper,we introduce a novel framework for knowledge-based VQA titled“Prompting Large Language Models with Knowledge-Injection”(PLLMKI).We use vanilla VQA model to inspire the LLM and further enhance the LLM with knowledge *** earlier approaches,we adopt the LLM for knowledge enhancement instead of relying on knowledge ***,we leverage open LLMs,incurring no additional *** comparison to existing baselines,our approach exhibits the accuracy improvement of over 1.3 and 1.7 on two knowledge-based VQA datasets,namely OK-VQA and A-OKVQA,respectively.

关键词： visual question answering knowledge-based visual question answering large language model knowledge injection

来源：评论

学校读者我要写书评

暂无评论

Innovative Approaches to Task Scheduling in Cloud Computing Environments Using an Advanced Willow Catkin Optimization Algorithm

引用

computers, Materials & Continua 2025年第2期82卷 2495-2520页

作者： Jeng-Shyang Pan Na Yu Shu-Chuan Chu An-Ning Zhang Bin Yan Junzo Watada College of Computer Science and Engineering Shandong University of Science and TechnologyQingdao266590China Department of Information Management Chaoyang University of TechnologyTaichung41349Taiwan College of Electronics Communication and PhysicsShandong University of Science and TechnologyQingdao266590China Graduate School of Information Production and SystemsWaseda UniversityKitakyushu808-0135Japan

The widespread adoption of cloud computing has underscored the critical importance of efficient resource allocation and management, particularly in task scheduling, which involves assigning tasks to computing resources for optimized resource utilization. Several meta-heuristic algorithms have shown effectiveness in task scheduling, among which the relatively recent Willow Catkin Optimization (WCO) algorithm has demonstrated potential, albeit with apparent needs for enhanced global search capability and convergence speed. To address these limitations of WCO in cloud computing task scheduling, this paper introduces an improved version termed the Advanced Willow Catkin Optimization (AWCO) algorithm. AWCO enhances the algorithm’s performance by augmenting its global search capability through a quasi-opposition-based learning strategy and accelerating its convergence speed via sinusoidal mapping. A comprehensive evaluation utilizing the CEC2014 benchmark suite, comprising 30 test functions, demonstrates that AWCO achieves superior optimization outcomes, surpassing conventional WCO and a range of established meta-heuristics. The proposed algorithm also considers trade-offs among the cost, makespan, and load balancing objectives. Experimental results of AWCO are compared with those obtained using the other meta-heuristics, illustrating that the proposed algorithm provides superior performance in task scheduling. The method offers a robust foundation for enhancing the utilization of cloud computing resources in the domain of task scheduling within a cloud computing environment.

关键词： Willow catkin optimization algorithm cloud computing task scheduling opposition-based learning strategy

来源：评论

学校读者我要写书评

暂无评论

UPMatch: Enhancing Semi-Supervised Medical Image Classification through Contrastive Learning with Unreliable Pseudo Labels 46

UPMatch: Enhancing Semi-Supervised Medical Image Classificat...

引用

46th Annual International Conference of the IEEE engineering in Medicine and Biology Society, EMBC 2024

作者： Hu, Jihong Li, Yinhao Lin, Lanfen Chen, Yen-Wei Ritsumeikan University The Graduate School of Information Science and Engineering Osaka Japan Ritsumeikan University The College of Information Science and Engineering Osaka Japan Zhejiang University The College of Computer Science and Technology Hangzhou China

ISBN: (纸本)9798350371499

Pseudo-labeling based semi-supervised learning (SSL) framework has proven highly successful in medical image analysis (MIA) by addressing the problem of a shortage of labeled samples. However, the existing SSL methods use a fixed or flexible confidence threshold to filter reliable samples, leaving large number of unlabeled samples unused. This is a more serious issue in MIA because of the low inter-class distance and imbalanced categories. We argue that effectively mining useful information hidden in ambiguous unlabeled sample is the key to improve model performance, so we propose UPmatch, a new pseudo labeling-based SSL framework. Our framework introduces a contrastive unreliable pseudo label learning module (CUPM) that incorporates unreliable pseudo label samples into the training process. Additionally, we propose an informative sample selection strategy (ISSS) that selects samples used in contrastive learning iteratively in each mini-batch. Our experiments on TissueMNIST and ISIC2019 dataset with various training settings demonstrate the effectiveness of our proposed strategy. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Jointprop: Joint Semi-supervised Learning for Entity and Relation Extraction with Heterogeneous Graph-based Propagation 61

Jointprop: Joint Semi-supervised Learning for Entity and Rel...

引用

61st Annual Meeting of the Association for Computational Linguistics, ACL 2023

作者： Yandan, Zheng Anran, Hao Tuan, Luu Anh School of Computer Science and Engineering Interdisciplinary Graduate Program-HealthTech Nanyang Technological University Singapore

ISBN: (纸本)9781959429722

Semi-supervised learning has been an important approach to address challenges in extracting entities and relations from limited data. However, current semi-supervised works handle the two tasks (i.e., Named Entity Recognition and Relation Extraction) separately and ignore the cross-correlation of entity and relation instances as well as the existence of similar instances across unlabeled data. To alleviate the issues, we propose Jointprop, a Heterogeneous Graph-based Propagation framework for joint semi-supervised entity and relation extraction, which captures the global structure information between individual tasks and exploits interactions within unlabeled data. Specifically, we construct a unified span-based heterogeneous graph from entity and relation candidates and propagate class labels based on confidence scores. We then employ a propagation learning scheme to leverage the affinities between labelled and unlabeled samples. Experiments on benchmark datasets show that our framework outperforms the state-of-the-art semi-supervised approaches on NER and RE tasks. We show that the joint semi-supervised learning of the two tasks benefits from their codependency and validates the importance of utilizing the shared information between unlabeled data. © 2023 Association for Computational Linguistics.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Intelligent Decision Support System for Higher Education Institutions 9

Intelligent Decision Support System for Higher Education Ins...

引用

9th International Conference on Signal Processing and Intelligent Systems, ICSPIS 2023

作者： Afriliana, Nunik Meyliana Gaol, Ford Lumban Soeparno, Haryono Bina Nusantara University Binus Graduate Program Computer Science Program Computer Science Department Jakarta Indonesia Bina Nusantara University School of Information System Information System Department Jakarta Indonesia

ISBN: (纸本)9798350369922

To ensure the sustainability of their operations, higher education institutions have to establish a robust strategic plan. Higher education institutions have an obligation to continuously assess the quality of their educational offerings in every program to comply with the standards established by the government and other educational entities. The implementation of intelligent decision support systems is a feasible approach among other techniques for addressing the aforementioned challenges. An intelligent decision support system has the potential to enable the business to uphold its adherence to regulations, enhance operational effectiveness, and optimize overall performance. This study aimed to determine the current state of intelligent decision support systems through the implementation of a comprehensive literature review. The second goal is to evaluate the best practices, methods, and implementation of intelligent decision support systems within higher education institutions. The findings indicate that intelligent decision support systems have the capacity to serve as a viable option for higher education, particularly in terms of enhancing the accuracy and efficiency of decision-making processes. Intelligent decision support systems exhibit a strong correlation with the fields of data mining and computational intelligence techniques. Nevertheless, the amount of research conducted on this subject is limited. © 2023 IEEE.

关键词： Data mining

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：