检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Li, Peize Si, Qingyi Fu, Peng Lin, Zheng Wang, Yan School of Artificial Intelligence Jilin University Changchun China Institute of Information Engineering Chinese Academy of Sciences Beijing China Huawei Technologies Co. Ltd. Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University Changchun China

ISBN: (纸本)157735897X

Retrieval-based multi-image question answering (QA) task involves retrieving multiple question-related images and synthesizing these images to generate an answer. Conventional "retrieve-then-answer" pipelines often suffer from cascading errors because the training objective of QA fails to optimize the retrieval stage. To address this issue, we propose a novel method to effectively introduce and reference retrieved information into the QA. Given the image set to be retrieved, we employ a multimodal large language model (visual perspective) and a large language model (textual perspective) to obtain multimodal hypothetical summary in question-form and description-form. By combining visual and textual perspectives, MHyS captures image content more specifically and replaces real images in retrieval, which eliminates the modality gap by transforming into text-to-text retrieval and helps improve retrieval. To more advantageously introduce retrieval with QA, we employ contrastive learning to align queries (questions) with MHyS. Moreover, we propose a coarse-to-fine strategy for calculating both sentence-level and word-level similarity scores, to further enhance retrieval and filter out irrelevant details. Our approach achieves a 3.7% absolute improvement over state-of-the-art methods on RETVQA and a 14.5% improvement over CLIP. Comprehensive experiments and detailed ablation studies demonstrate the superiority of our method. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

FISS GAN:A Generative Adversarial Network for Foggy Image Semantic Segmentation

引用

IEEE/CAA Journal of Automatica Sinica 2021年第8期8卷 1428-1439页

作者： Kunhua Liu Zihao Ye Hongyan Guo Dongpu Cao Long Chen Fei-Yue Wang School of Computer Science and Engineering Sun Yat-sen UniversityGuangzhou 510006China State Key Laboratory of Automotive Simulation and Control and the Department of Control Science and Engineering Jilin University(Campus Nanling)Changchun 130025China Mechanical and Mechatronics Engineering Department at the University of Waterloo 200 University Avenue West WaterlooON N2L 3G1Canada State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190 Institute of Systems Engineering Macao University of Science and TechnologyMacao 999078China

Because pixel values of foggy images are irregularly higher than those of images captured in normal weather(clear images),it is difficult to extract and express their *** method has previously been developed to directly explore the relationship between foggy images and semantic segmentation *** investigated this relationship and propose a generative adversarial network(GAN)for foggy image semantic segmentation(FISS GAN),which contains two parts:an edge GAN and a semantic segmentation *** edge GAN is designed to generate edge information from foggy images to provide auxiliary information to the semantic segmentation *** semantic segmentation GAN is designed to extract and express the texture of foggy images and generate semantic segmentation *** on foggy cityscapes datasets and foggy driving datasets indicated that FISS GAN achieved state-of-the-art performance.

关键词： Edge GAN foggy images foggy image semantic segmentation GAN semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Chinese Clinical Named Entity Recognition Based on Stroke-Level and Radical-Level Features 1

引用

5th International Conference on Smart Computing and Communication, SmartCom 2020

作者： Zhou, Feng Han, Xuming Liu, Qiaoming Li, Mingyang Li, Yong School of Computer Science and Engineering Changchun University of Technology Changchun130012 China College of Information Science and Technology Jinan University Guangzhou510632 China School of Computer Science and Technology Harbin Institute of Technology Harbin150006 China School of Management Jilin University Changchun130022 China

ISBN: (数字)9783030747176

ISBN: (纸本)9783030747169

Clinical Named Entity Recognition (CNER) is an important step for mining clini-cal text. Aiming at the problem of insufficient representation of potential Chinese features, we propose the Chinese clinical named entity recognition model based on stroke level and radical level features. The model leverages Bidirectional Long Short-term Memory (BiLSTM) neural network to extract the internal semantic in-formation of Chinese characters (i.e., strokes and radicals). Our method can not only capture the dependence of the internal strokes of Chinese characters, but also enhance the semantic representation of Chinese characters, thereby improving the entity recognition ability of the model. Experimental results show that the accuracy of the model on the CCKS-2017 task 2 benchmark data set reaches 93.66%, and the F1 score reaches 94.70%. Comp ared with the basic BiLSTM-CRF mod-el, the precision of model is increased by 3.38%, the recall is increased by 1.05% and F1 value is increased by 1.91%. © 2021, Springer Nature Switzerland AG.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Transmission Line Fault Ranging Method Based on ICEEMDAN-HFWEO

Transmission Line Fault Ranging Method Based on ICEEMDAN-HFW...

引用

Cyber-Energy Systems and Intelligent Energy (ICCSIE), International Conference on

作者： Zhijie Song Linlin Li Gang Liu Zhiyong Yang Jiyao Zhang Minzhen Wang School of electrical and information engineering Changchun institute of technology Changchun China Da'an power supply center State grid jilin electric power company limited baicheng power supply company Baicheng China Substation Operation and Inspection Center State Grid Songyuan Power Supply Company and Engineering Songyuan China Research office Changchun university of architecture and Engineering Changchun China School of electronic information College of science and technology ning bo university Ningbo China

ISBN: (数字)9798350373318

ISBN: (纸本)9798350373325

In the traveling wave fault localization method, the fault traveling wave head extraction accuracy determines the fault localization accuracy, this paper proposes a transmission line fault ranging method based on ICEEMDAN-HFWEO as a way to improve the extraction accuracy of traveling wave head. Firstly, for the defect that the single line mode component of the traditional phase mode transform matrix cannot characterize all types of faults, the Clark phase mode transform matrix is improved so that the single line mode component can characterize all types of faults; on this basis, ICEEMDAN (Improved Complete Ensemble Empirical Mode Based on this, ICEEMDAN (Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) and HFWEO (Higher-order Frequency Weighted Energy Operator) are proposed as a combination of travelling waveform header extraction methods, which improves the extraction accuracy of the faulty travelling waveheads and realizes the accurate fault localization. Simulation experiments have proved that the proposed method can realize accurate fault localization of transmission lines under different transition resistances, different fault types and different fault distances, and the localization errors are less than 150m.

关键词： Location awareness Accuracy Power transmission lines Transmission line matrix methods Empirical mode decomposition Noise Distance measurement Matrix decomposition

来源：评论

学校读者我要写书评

暂无评论

High-Sensitivity Photoacoustic Gas Sensor With Ultrawide Dynamic Range Using Active Frequency Tracking Technology

引用

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Li, Yafei Guan, Gangyun Yang, Shuo Deng, Zhuobei Jiang, Yanyu Deng, Wanling Huang, Junkai Wang, Xicheng Xue, Xiaobin Han, Xile Zheng, Chuantao Guo, Tuan Tittel, Frank K. Jinan University College of Information Science and Technology Department of Electronic Engineering Guangzhou510630 China Jilin University State Key Laboratory of Integrated Optoelectronics College of Electronic Science and Engineering Changchun130012 China Jinan University Institute of Photonics Technology College of Physics and Optoelectronic Engineering Guangzhou511443 China Rice University Department of Electrical and Computer Engineering HoustonTX77005 United States

To achieve highly sensitive and ultrawide dynamic range gas detection, a dual-mode gas sensor was proposed using cavity-enhanced photoacoustic spectroscopy (CEPAS) and active resonant frequency tracking (ARFT) technology. CEPA cell with double resonance of optical and acoustic can greatly enhance the detection sensitivity of the sensor. Moreover, a new sensing mechanism based on the relationship between gas concentration and acoustic resonant frequency was studied. As an example, we developed an acetylene (C2H2) sensor, achieving a normalized noise equivalent absorption (NNEA) coefficient of 9:86 × 10-11 cm-1 W Hz-1/2 and an unprecedented dynamic range of nearly nine orders of magnitude (10-3 - 106 ppm). This new measurement tool will provide crucial additional capabilities to gas monitoring with improved sensitivity and dynamic range. © 2025 IEEE.

关键词： Gas detectors

来源：评论

学校读者我要写书评

暂无评论

StripeRust-Pocket:A Mobile-Based Deep Learning Application for Efficient Disease Severity Assessment of Wheat Stripe Rust

引用

Plant Phenomics 2024年第3期6卷 771-786页

作者： Weizhen Liu Yuxi Chen Zhaoxin Lu Xiaoyu Lu Ze Wu Ziyao Zheng Yongqiang Suo Caixia Lan Xiaohui Yuan School of Computer Science and Artificial Intelligence Wuhan University of TechnologyWuhanHubei 430070China Wuhan University of Technology Chongqing Research Institute Chongqing 401120China Sanya Science and Education Innovation Park of Wuhan University of Technology SanyaHainan 572025China Hubei Hongshan Laboratory College of Plant Science and TechnologyHuazhong Agricultural UniversityWuhan 430070HubeiChina Engineering Research Centre of Chinese Ministry of Education for Edible and Medicinal Fungi Jilin Agricultural UniversityChangchunJilin 130118China

Wheat stripe rust poses a marked threat to global wheat *** and effective disease severity assessments are crucial for disease resistance breeding and timely management of field *** this study,we propose a practical solution using mobile-based deep learning and model-assisted ***-Pocket,a user-friendly mobile application developed based on deep learning models,accurately quantifies disease severity in wheat stripe rust leaf images,even under complex ***,StripeRust-Pocket facilitates image acquisition,result storage,organization,and *** underlying model employed by StripeRust-Pocket,called StripeRustNet,is a balanced lightweight 2-stage *** first stage utilizes MobileNetV2-DeepLabV3+for leaf segmentation,followed by ResNet50-DeepLabV3+in the second stage for lesion *** severity is estimated by calculating the ratio of the lesion pixel area to the leaf pixel *** achieves 98.65%mean intersection over union(MIoU)for leaf segmentation and 86.08%MIoU for lesion *** using an additional 100 field images demonstrated a mean correlation of over 0.964 with 3 expert visual *** address the challenges in manual labeling,we introduce a 2-stage labeling pipeline that combines model-assisted labeling,manual correction,and spatial *** apply this pipeline to our self-collected dataset,reducing the annotation time from 20 min to 3 min per *** method provides an efficient and practical solution for wheat stripe rust severity assessments,empowering wheat breeders and pathologists to implement timely disease *** also demonstrates how to address the"last mile"challenge of applying computer vision technology to plant phenomics.

关键词： application learning pocket deep assessment disease efficient mobile-based severity striperust

来源：评论

学校读者我要写书评

暂无评论

Aqua-Sim Fourth Generation: Towards General and Intelligent Simulation for Underwater Acoustic Networks

arXiv

引用

arXiv 2024年

作者： Guo, Jiani Song, Shanshan Chen, Hao Huangfu, Bingwen Liu, Jun Cui, Jun-Hong The College of Computer Science and Technology Jilin University Changchun130012 China The School of Electronic and Information Engineering Beihang University Beijing100191 China The Shenzhen Ocean Information Technology Industry Research Institute Shenzhen518055 China

Simulators are essential to troubleshoot and optimize Underwater Acoustic Network (UAN) schemes (network protocols and communication technologies) before real field experiments. However, due to programming differences between the above two contents, most existing simulators concentrate on one while weakening the other, leading to non-generic simulations and biased performance results. Moreover, novel UAN schemes increasingly integrate Artificial Intelligence (AI) techniques, yet existing simulators lack support for necessary AI frameworks, failing to train and evaluate these intelligent methods. On the other hand, these novel schemes consider more UAN characteristics involving more complex parameter configurations, which also challenges simulators in flexibility and fineness. To keep abreast of advances in UANs, we propose the Fourth Generation (FG) ns-3-based simulator Aqua-Sim FG, enhancing the general and intelligent simulation ability. On the basis of retaining previous generations’ functions, we design a new general architecture, which is compatible with various programming languages, including MATLAB, C++, and Python. In this way, Aqua-Sim FG provides a general environment to simulate communication technologies, network protocols, and AI models simultaneously. In addition, we expand six new features from node and communication levels by considering the latest UAN methods’ requirements, which enhances the simulation flexibility and fineness of Aqua-Sim FG. Experimental results show that Aqua-Sim FG can simulate UANs’ performance realistically, reflect intelligent methods’ problems in real-ocean scenarios, and provide more effective troubleshooting and optimization for actual UANs. The basic simulator is available at https://***/JLU-smartocean/aqua-sim-fg. Copyright © 2024, The Authors. All rights reserved.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Chlorine-substituted chiral 2D perovskite microwires with enhanced lattice rigidity for sensitive circularly polarized light detection

引用

science China Chemistry 2023年第12期66卷 3602-3610页

作者： Meng Yuan Yuchen Qiu Yingjie Zhao Yuyan Zhao Hui Li Xiao Wei Gaosong Chen Jiangang Feng Hanfei Gao Jinjin Zhao Jiahui Zhao Lei Jiang Yuchen Wu Key Laboratory of Bio-inspired Materials and Interfacial Science Technical Institute of Physics and ChemistryChinese Academy of SciencesBeijing 100190China College of Chemistry Zhengzhou UniversityZhengzhou 450001China College of Chemistry Jilin UniversityChangchun 130012China Key Laboratory for Special Functional Materials of Ministry of Education National and Local Joint Engineering Research Center for High-Efficiency Display and Lighting TechnologyCollaborative Innovation Center of Nano Functional Materials and ApplicationsHenan UniversityKaifeng 475004China University of Chinese Academy of Sciences(UCAS) Beijing 100049China Department of Electrical and Computer Engineering National University of SingaporeSingapore 117585Singapore Suzhou Institute for Advanced Research University of Science and Technology of ChinaSuzhou 215123China Ji Hua Laboratory Foshan 528000China Department of Physics Shanxi Datong UniversityDatong 037009China

Chiral two-dimensional(2D) perovskites offer numerous attractive features for optoelectronics owing to their soft, deformable lattices and a high degree of chemical tunability. While tremendous advances have been made in perovskite-based direct circularly polarized light(CPL) detection, the low circular polarization anisotropy factor and sensitivity of those photodetectors arising from large lattice distortion still limit practical applications. Herein, chiral 2D perovskite-based single-crystalline microwire arrays with enhanced circular dichroism(CD) absorption are fabricated with the synergy of the capillary-bridgeconfined assembly method and chlorine-substituted phenethylamine(Cl-MBA). Compared with phenethylamine(MBA)-inserted perovskites, the smaller lattice distortion and increased halogen–halogen interaction within Cl–MBA-inserted perovskites strengthen lattice rigidity and weaken electron–phonon coupling to improve carrier transport and thermal stability, resulting in high-performance CPL photodetectors with an anisotropy factor of 0.25, responsivity exceeding 95.7±9.3 A W^(-1)and detectivity exceeding(3.05±0.30)×10^(13)Jones. This work opens a new perspective to modulate circular polarization sensitivity and will be helpful to realize promising implementations in quantum computation and communication.

关键词： chiral perovskites microwire arrays lattice rigidity circular polarization photodetectors

来源：评论

学校读者我要写书评

暂无评论

A Multi-Task-Based Deep Multi-Scale Information Fusion Method for Intelligent Diagnosis of Bearing Faults

SSRN

引用

SSRN 2022年

作者： Xin, Ruihao Wang, Tiantian Miao, Fengbo Yu, Cuinan Feng, Xin College of Information and Control Engineering Jilin Institute of Chemical Technology Jilin Jilin132000 China College of Computer Science and Technology Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Jilin Changchun130012 China School of Science Jilin Institute of Chemical Technology Jilin Jilin132000 China

The use of deep learning for fault diagnosis is already a commonly used method. However, complex operating conditions under different working conditions significantly impact bearing vibration signals. Integrating the discriminative information of fault type and size into the deep learning model to achieve rich multi-task fault feature diagnosis still deserves attention. This study proposes a deep multi-task-based multi-scale feature fusion network MEAT(multi-task-based deep multi-scale feature fusion network) model to address these issues. Solve the limitations and poor adaptability of traditional convolutional neural network models under complex work. This model considers two sub-tasks, fault scale discrimination, and fault type discrimination, in the fault diagnosis multi-classification task through multi-task mapping decomposition. First, multi-dimensional features are extracted by convolution at different scales to obtain multi-dimensional fault information at different levels. Then these features are weighted and fused using a multi-layer attention mechanism to get accurate diagnosis results of bearing faults. © 2022, The Authors. All rights reserved.

关键词： Failure analysis

来源：评论

学校读者我要写书评

暂无评论

Multimodal Hypothetical Summary for Retrieval-based Multi-image Question Answering

arXiv

引用

arXiv 2024年

作者： Li, Peize Si, Qingyi Fu, Peng Lin, Zheng Wang, Yan School of Artificial Intelligence Jilin University Changchun China Institute of Information Engineering Chinese Academy of Sciences Beijing China Huawei Technologies Co. Ltd. Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Symbol Computation and Knowledge Engineering Ministry of Education College of Computer Science and Technology Jilin University Changchun China

Retrieval-based multi-image question answering (QA) task involves retrieving multiple question-related images and synthesizing these images to generate an answer. Conventional "retrieve-then-answer" pipelines often suffer from cascading errors because the training objective of QA fails to optimize the retrieval stage. To address this issue, we propose a novel method to effectively introduce and reference retrieved information into the QA. Given the image set to be retrieved, we employ a multimodal large language model (visual perspective) and a large language model (textual perspective) to obtain multimodal hypothetical summary in question-form and description-form. By combining visual and textual perspectives, MHyS captures image content more specifically and replaces real images in retrieval, which eliminates the modality gap by transforming into text-to-text retrieval and helps improve retrieval. To more advantageously introduce retrieval with QA, we employ contrastive learning to align queries (questions) with MHyS. Moreover, we propose a coarse-to-fine strategy for calculating both sentence-level and word-level similarity scores, to further enhance retrieval and filter out irrelevant details. Our approach achieves a 3.7% absolute improvement over state-of-the-art methods on RETVQA and a 14.5% improvement over CLIP. Comprehensive experiments and detailed ablation studies demonstrate the superiority of our method. Copyright © 2024, The Authors. All rights reserved.

关键词： Question answering

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：