检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Ma, Mingjie Yu, Zhihuan Ma, Yichao Li, Guohui Yang, Zhong School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China School of Software Engineering Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798350368741

Visual Commonsense Reasoning (VCR) is a cognitive task, challenging models to answer visual questions, and to explain the rationale behind their answers. While Large Language Models (LLMs) offer potential for this task, VCR's complex scenes require specialized approaches to activate their commonsense reasoning abilities, as existing Multimodal LLMs struggle with VCR's visual events and unique reference tags. To address these challenges, we propose EventLens, which enhances VCR through Event-Aware Pretraining and Cross-modal Linking in Supervised Fine-tuning. First, we introduce a new pretraining stage that emulates human cognitive processes to improve LLM comprehension of complex scenarios. Second, during supervised fine-tuning, we leverage reference tags to explicitly bridge RoI features with text, maintaining semantic integrity across modalities. Additionally, instruct prompts and task-specific adapters help integrate LLMs' knowledge with new commonsense reasoning. Experimental results demonstrate competitive performance with state-of-the-art methods and ablation studies verify the effectiveness of proposed EventLens. © 2025 IEEE.

关键词： Multimedia Understanding Pretrained Language Models Visual Commonsense Reasoning

来源：评论

学校读者我要写书评

暂无评论

Streaming Histogram Publication over Weighted Sliding Windows Under Differential Privacy

引用

Tsinghua science and Technology 2024年第6期29卷 1674-1693页

作者： Xiujun Wang Lei Mo Xiao Zheng Zhe Dang School of Computer Science and Technology Anhui University of TechnologyMa’anshan 243032China Institute for Artificial Intelligence Hefei Comprehensive National Science CenterHefei 230088China Baosight Software(Anhui)Co.Ltd. Ma’anshan 243000China School of Electrical Engineering and Computer Science Washington State UniversityPullman 99164WAUSA

Continuously publishing histograms in data streams is crucial to many real-time applications,as it provides not only critical statistical information,but also reduces privacy leaking *** the importance of elements usually decreases over time in data streams,in this paper we model a data stream by a sequence of weighted sliding windows,and then study how to publish histograms over these windows *** existing literature can hardly solve this problem in a real-time way,because they need to buffer all elements in each sliding window,resulting in high computational overhead and prohibitive storage *** this paper,we overcome this drawback by proposing an online algorithm denoted by Efficient Streaming Histogram Publishing(ESHP)to continuously publish histograms over weighted sliding ***,our method first creates a novel sketching structure,called Approximate-Estimate Sketch(AESketch),to maintain the counting information of each histogram interval at every time instance;then,it creates histograms that satisfy the differential privacy requirement by smartly adding appropriate noise values into the sketching *** experimental results and rigorous theoretical analysis demonstrate that the ESHP method can offer equivalent data utility with significantly lower computational overhead and storage costs when compared to other existing methods.

关键词： differential privacy randomized algorithm streaming data publication weighted sliding window approximate statistics data usability computational complexity

来源：评论

学校读者我要写书评

暂无评论

MILG:Realistic lip-sync video generation with audio-modulated image inpainting

引用

Visual Informatics 2024年第3期8卷 71-81页

作者： Han Bao Xuhong Zhang Qinying Wang Kangming Liang Zonghui Wang Shouling Ji Wenzhi Chen School of Software Technology Zhejiang UniversityHangzhouChina College of Computer Science Zhejiang UniversityHangzhouChina College of Engineering Zhejiang UniversityHangzhouChina

Existing lip synchronization(lip-sync)methods generate accurately synchronized mouths and faces in a generated ***,they still confront the problem of artifacts in regions of non-interest(RONI),e.g.,background and other parts of a face,which decreases the overall visual *** solve these problems,we innovatively introduce diverse image inpainting to lip-sync *** propose Modulated Inpainting Lip-sync GAN(MILG),an audio-constraint inpainting network to predict synchronous *** utilizes prior knowledge of RONI and audio sequences to predict lip shape instead of image generation,which can keep the RONI ***,we integrate modulated spatially probabilistic diversity normalization(MSPD Norm)in our inpainting network,which helps the network generate fine-grained diverse mouth movements guided by the continuous audio ***,to lower the training overhead,we modify the contrastive loss in lipsync to support small-batch-size and few-sample *** experiments demonstrate that our approach outperforms the existing state-of-the-art of image quality and authenticity while keeping lip-sync.

关键词： Lip-sync Image inpainting Face generation Modulated SPD normalization

来源：评论

学校读者我要写书评

暂无评论

Enhancing Ovarian Cancer Diagnosis Through Deep Learning: A VGG19-Based Approach for Subtype Identification

Enhancing Ovarian Cancer Diagnosis Through Deep Learning: A ...

引用

2024 OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Agarwal, Muskan Aggarwal, Priyanshi Sunil, G. Gill, Kanwarpartap Singh Rawat, Ramesh Singh Chitkara University Computer Science & Engineering Punjab India Graphic Era Hill University Computer Science & Engineering Uttarakhand Dehradun248002 India Sr University School of Computer Science & Artificial Intelligence Warangal India Graphic Era Deemed to Be University Computer Science& Engineering Uttarakhand Dehradun248002 India

ISBN: (纸本)9798350373783

The abstract of this research paper outlines the overarching objectives, methodologies, findings, and implications of the study, focusing on the development of a robust machine learning model for precise classification of various ovarian cancer subtypes. The primary goal was to leverage the VGG19 architecture, a deep convolutional neural network, to accurately identify subtypes including endometrioid carcinoma (EC), clear cell carcinoma (CC), mucinous carcinoma (MC), high-grade serous carcinoma (HGSC), and low-grade serous carcinoma (LGSC). Employing advanced image processing algorithms and feature extraction techniques, the model was trained on a dataset comprising histology images of ovarian tissue samples. The achieved overall accuracy of 75% signifies the effectiveness of the proposed strategy in subtype identification. A comprehensive evaluation of the model's performance, including accuracy, recall, and F1-score, was conducted to assess its classification skills across diverse subtypes. The study not only presents the model's achievements but also sheds light on potential areas for growth, future research prospects, and an analysis of the model's strengths and weaknesses. The research contributes significantly to the field by demonstrating the ability to differentiate between ovarian cancer subgroups, offering valuable insights that could enhance the precision of diagnosis and, consequently, the formulation of tailored treatment regimens. The abstract concludes by emphasizing the increasing trend in the utilization of computational approaches, particularly machine learning, for enhanced accuracy in ovarian cancer diagnosis and treatment. Overall, the abstract serves as a concise and informative preview of the research, inviting readers to delve deeper into the study's details. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

General or Specific? Retrofitting AI and IoT-Enabled Hydroponic Grow Tents for Use in Different Climates Across Rural Africa 4

General or Specific? Retrofitting AI and IoT-Enabled Hydropo...

引用

4th International Conference on Emerging Trends in Networks and computer Communications, ETNCC 2024

作者： Till, Sarina Wilson, Taryn Karamata, Rikurora Beukes, Franco Kaundje, Metaramuje Winschiers-Theophilus, Heike School of Computer Science IIE Durban South Africa Software Engineering NUST Windhoek Namibia

ISBN: (纸本)9798350353266

Climate change threatens African subsistence farming with erratic rainfall, droughts, and seasonal shifts causing devastating crop failures. Hydroponic farming, often recommended to mitigate these threats, requires constant monitoring and knowledge, which can be overwhelming for first-time growers. Research has explored using the Internet of Things (IoT) and Artificial Intelligence (AI) to automate and monitor hydroponic systems. Still, few studies have tested these systems in the field with farmers. Our previous work involved an eight-month deployment of an AI and IoT-enabled hydroponic system with rural subsistence farmers in South Africa. We conducted focus groups and observed the hydroponic systems using a qualitative approach. This paper discusses lessons learned from deploying the system in Namibia. We found that generalized AI models are unlikely to succeed due to differences in water quality, climate, participant readiness, and internet connectivity. Therefore, AI and IoT systems for rural Africa must be tailor-made to be functional and effective for crop growth. © 2024 IEEE.

关键词： Hydroponics

来源：评论

学校读者我要写书评

暂无评论

Dynamic Countdown Scheduling MAC Protocol in Underwater Acoustic Sensor Networks

引用

IEEE Sensors Journal 2025年第11期25卷 20797-20812页

作者： Feng, Libin Li, Zhe Mai, Haohao Chen, Yanxia Huang, Xiangdang Hainan University School of Computer Science and Technology Haikou570228 China Hainan University School of Software Engineering Haikou570228 China Hainan University School of Cyberspace Security Haikou570228 China Hainan University and the Innovation Platform for Academicians of Hainan Province School of Computer Science and Technology Haikou570228 China

Underwater acoustic sensor networks (UASNs) are highly sensitive to collisions due to the unique characteristics of underwater acoustic signal propagation. Thus, developing medium access control (MAC) protocols that can effectively avoid collisions has become particularly important. This paper proposes a dynamic countdown scheduling MAC protocol (DCS-MAC) to achieve collision-free transmission for UASNs. First, transmission schedules of nodes are integrated directly into data frames, eliminating the need for control frames, thereby reducing latency and energy consumption. Second, considering defined collision constraint, a dynamic countdown strategy is introduced where the transmission schedule of each node dynamically adjusts according to collected collision constraints, reducing the probability of data frame collisions. Then, interval trees are used to dynamically update collision constraints, maximizing the coverage of collision time periods and selecting the optimal transmission schedule. Finally, simulation results show that DCS-MAC excels in data frame delivery rate, network throughput, throughput fairness, and energy consumption. © 2001-2012 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection in Spatial Information Networks: A Security Knowledge Graph-based Approach

Anomaly Detection in Spatial Information Networks: A Securit...

引用

2024 International Conference on Intelligent Computing and Robotics, ICICR 2024

作者： Tan, Yayue Zhang, Yaqi Liu, Hui Tang, Yongqing College of Computer and Information Science School of Software Southwest University Chongqing China School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350372274

Anomaly detection stands as a critical element in securing space information networks (SINs). This paper delves into the realm of anomaly detection within dynamic networks, shedding light on established methodologies. The focus lies on scrutinizing various anomaly detection strategies in space information networks, conducting a comprehensive assessment of their practical applications, advantages, and drawbacks. Moreover, it introduces a novel anomaly detection approach driven by deep learning principles. This innovative method merges graph convolution techniques with a network securityoriented knowledge graph, presenting a cutting-edge solution. As a final contribution, the paper extends its purview to the future landscape of anomaly detection within space information networks, offering valuable insights into potential advancements and forthcoming trends in this domain. This holistic investigation seeks to enhance the understanding of anomaly detection paradigms within the unique context of space information networks, aiming to fortify their security measures and operational resilience. ©2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

ELTA: An Enhancer against Long-Tail for Aesthetics-oriented Models 41

ELTA: An Enhancer against Long-Tail for Aesthetics-oriented ...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Liu, Limin He, Shuai Ming, Anlong Xie, Rui Ma, Huadong School of Computer Science National Pilot Software Engineering School Beijing University of Posts and Telecommunications Beijing China

Real-world datasets often exhibit long-tailed distributions, compromising the generalization and fairness of learning-based models. This issue is particularly pronounced in Image Aesthetics Assessment (IAA) tasks, where such imbalance is difficult to mitigate due to a severe distribution mismatch between features and labels, as well as the great sensitivity of aesthetics to image variations. To address these issues, we propose an Enhancer against Long-Tail for Aesthetics-oriented models (ELTA). ELTA first utilizes a dedicated mixup technique to enhance minority feature representation in high-level space while preserving their intrinsic aesthetic qualities. Next, it aligns features and labels through a similarity consistency approach, effectively alleviating the distribution mismatch. Finally, ELTA adopts a specific strategy to refine the output distribution, thereby enhancing the quality of pseudo-labels. Experiments on four representative datasets (AVA, AADB, TAD66K, and PARA) show that our proposed ELTA achieves state-of-the-art performance by effectively mitigating the long-tailed issue in IAA datasets. Moreover, ELTA is designed with plug-and-play capabilities for seamless integration with existing methods. To our knowledge, this is the first contribution in the IAA community addressing long-tail. All resources are available in here. Copyright 2024 by the author(s)

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

A Social Distancing Framework Based on GPS and Bluetooth Empowered by Feature-based Machine Learning Algorithm

引用

IEIE Transactions on Smart Processing and Computing 2025年第1期14卷 118-127页

作者： Khan, Muhammad Adnan Rehman, Abdur Abbas, Sagheer Ali, Muhammad Nadeem Kim, Byung-Seo Department of Software Faculty of Artificial Intelligence & Software Gachon University Seongnam13120 Korea Republic of School of Computer Science National College of Business Administration and Economics Lahore54000 Pakistan Department of Computer Science Prince Mohammad Bin Fahd University Al Khobar Saudi Arabia Department of Software and Communications Engineering Hongik University Sejong Campus Korea Republic of

The current COVID-19 epidemic is responsible for causing a catastrophe on a global scale due to its risky spread. The community’s insecurity is growing as a result of a lack of appropriate remedial measures and immunization against the disease. In this case, social distancing is thought to be an effective barrier against the spread of the contagion virus as the risk of virus transmission can be reduced by avoiding direct contact with people. Thus, the goal of this research is to develop and improve an AI (Artificial Intelligence) system architecture for social distance monitoring. The framework could also use the GPS (Global Positioning System) to recognize human separation through cell phones. The transition learning framework is also applied to improve the consistency of the existing system. In this manner, the detection system uses a pre-trained technique that takes a Bluetooth dataset and location-sharing dataset to link to an additional level. In an attempt to approximate social distancing breaches among people, we used Bluetooth technology along with GPS distance estimation and set a threshold. To predict if the distance value exceeds the required social distance standard, a violation threshold is calculated and then it sends an alarm to every individual who is not maintaining social distancing. In response, the individual who breaks the social distance limit is also monitored using a detection approach. © 2025 Institute of Electronics Engineers of Korea. All rights reserved.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

How far are we to GPT-4V?Closing the gap to commercial multimodal models with open-source suites

引用

science China(Information sciences) 2024年第12期67卷 5-22页

作者： Zhe CHEN Weiyun WANG Hao TIAN Shenglong YE Zhangwei GAO Erfei CUI Wenwen TONG Kongzhi HU Jiapeng LUO Zheng MA Ji MA Jiaqi WANG Xiaoyi DONG Hang YAN Hewei GUO Conghui HE Botian SHI Zhenjiang JIN Chao XU Bin WANG Xingjian WEI Wei LI Wenjian ZHANG Bo ZHANG Pinlong CAI Licheng WEN Xiangchao YAN Min DOU Lewei LU Xizhou ZHU Tong LU Dahua LIN Yu QIAO Jifeng DAI Wenhai WANG State Key Laboratory for Novel Software Technology Nanjing University Shanghai AI Laboratory School of Computer Science Fudan University SenseTime Research Department of Information Engineering The Chinese University of Hong Kong Department of Electronic Engineering Tsinghua University

In this paper, we introduce InternVL 1.5, an open-source multimodal large language model(MLLM) to bridge the capability gap between open-source and proprietary commercial models in multimodal understanding. We introduce three simple improvements.(1) Strong vision encoder: we explored a continuous learning strategy for the large-scale vision foundation model — InternViT-6B, boosting its visual understanding capabilities, and making it can be transferred and reused in different LLMs.(2) Dynamic high-resolution: we divide images into tiles ranging from 1 to 40 of 448×448 pixels according to the aspect ratio and resolution of the input images, which supports up to 4K resolution input.(3) High-quality bilingual dataset: we carefully collected a high-quality bilingual dataset that covers common scenes, document images,and annotated them with English and Chinese question-answer pairs, significantly enhancing performance in optical character recognition(OCR) and Chinese-related tasks. We evaluate InternVL 1.5 through a series of benchmarks and comparative studies. Compared to both open-source and proprietary commercial models, InternVL 1.5 shows competitive performance, achieving state-of-the-art results in 8 of 18 multimodal benchmarks. Code and models are available at https://***/OpenGVLab/InternVL.

关键词： multimodal model open-source vision encoder dynamic resolution bilingual dataset

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：