检索结果-内蒙古大学图书馆

2022 European Wireless Conference, EW 2022

作者： Zhu, Yao Hu, Yulin Tong, Wuwan Caire, Giuseppe Schmeink, Anke Information Theory and Data Analytics RWTH Aachen University AachenD-52074 Germany Faculty of Electrical Engineering and Computer Science Technical University of Berlin Berlin10587 Germany

ISBN: (纸本)9781713865698

Joint communication and sensing (JCAS) is envisioned as the enabler to many advanced applications in future 6G networks. This paper investigates a power allocation strategy for JCAS networks with multiple transmitters and receivers. To take advantage of spatial diversity, the dual-functional radio signal is constructed via superimposed NOMA. Moreover, to fulfill the stringent latency requirement, transmissions are carried out with finite blocklength (FBL) codes. We formulate the optimization problem aimed at maximizing the minimal transmission rate among NOMA users while guaranteeing the target estimation requirement, which is represented by the Cramer-Rao lower bound (CRLB). An efficient approach is proposed to obtain the optimal solutions based on the characterization of communication and sensing metrics. Via simulation, the analytical finding are validated and insights into the trade-off between communication and radar sensing in JCAS networks are provided. © VDE VERLAG GMBH · Berlin · Offenbach.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

First Token Probability Guided RAG for Telecom Question Answering

arXiv

引用

arXiv 2025年

作者： Chen, Tingwei Chen, Jiayi Zhao, Zijian Chen, Haolong Zhang, Liang Zhu, Guangxu Shenzhen Research Institute of Big Data Shenzhen China Shenzhen China School of Computer Science and Engineering Sun Yat-sen University Guangzhou China

Large Language Models (LLMs) have garnered significant attention for their impressive general-purpose capabilities. For applications requiring intricate domain knowledge, Retrieval-Augmented Generation (RAG) has shown a distinct advantage in incorporating domain-specific information into LLMs. However, existing RAG research has not fully addressed the challenges of Multiple Choice Question Answering (MCQA) in telecommunications, particularly in terms of retrieval quality and mitigating hallucinations. To tackle these challenges, we propose a novel first token probability guided RAG framework. This framework leverages confidence scores to optimize key hyperparameters, such as chunk number and chunk window size, while dynamically adjusting the context. Our method starts by retrieving the most relevant chunks and generates a single token as the potential answer. The probabilities of all options are then normalized to serve as confidence scores, which guide the dynamic adjustment of the context. By iteratively optimizing the hyperparameters based on these confidence scores, we can continuously improve RAG performance. We conducted experiments to validate the effectiveness of our framework, demonstrating its potential to enhance accuracy in domain-specific MCQA tasks. Copyright © 2025, 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

Boundary-Aware Temporal Dynamic Pseudo-Supervision Pairs Generation for Zero-Shot Natural Language Video Localization 39

Boundary-Aware Temporal Dynamic Pseudo-Supervision Pairs Gen...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Deng, Xiongwen Tang, Haoyu Jiang, Han Zheng, Qinghai Zhu, Jihua School of Software Engineering Xi’an Jiaotong University China School of Software Shandong University China College of Computer and Data Science Fuzhou University China

ISBN: (纸本)157735897X

Zero-shot Natural Language Video Localization (NLVL) aims to automatically generate moments and corresponding pseudo queries from raw videos for the training of the localization model without any manual annotations. Existing approaches typically produce pseudo queries as simple words, which overlook the complexity of queries in real-world scenarios. Considering the powerful text modeling capabilities of large language models (LLMs), leveraging LLMs to generate complete queries that are closer to human descriptions is a potential solution. However, directly integrating LLMs into existing approaches introduces several issues, including insensitivity, isolation, and lack of regulation, which prevent the full exploitation of LLMs to enhance zero-shot NLVL performance. To address these issues, we propose BTDP, an innovative framework for Boundary-aware Temporal Dynamic Pseudo-supervision pairs generation. Our method contains two crucial operations: 1) Boundary Segmentation that identifies both visual boundaries and semantic boundaries to generate the atomic segments and activity descriptions, tackling the issue of insensitivity. 2) Context Aggregation that employs the LLMs with a self-evaluation process to aggregate and summarize global video information for optimized pseudo moment-query pairs, tackling the issue of isolation and lack of regulation. Comprehensive experimental results on the Charades-STA and ActivityNet Captions datasets demonstrate the effectiveness of our BTDP method. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Clustering-Based Color Reordering Method for Reversible data Hiding in Palette Images

引用

IET Image Processing 2025年第1期19卷

作者： Deng, Jianxuan Chen, Yi Wang, Hongxia Guo, Chun Cui, Yunhe Shen, Guowei State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang China Engineering Research Center of Text Computing & Cognitive Intelligence Ministry of Education Guizhou University Guiyang China School of Cyber Science and Engineering Sichuan University Chengdu China

A recent research work pointed out that the reversible data hiding algorithms proposed for gray-scale images can be implemented on the reconstructed palette images to improve embedding capacity and visual quality by reordering the color table. However, the reordering effect has a significant impact on performance improvement. Therefore, we propose a clustering-based color reordering method for reversible data hiding in palette images to improve the reordering effect and further enhance the performance. In this method, we first design a centroid initialization method to select the initial centroids and then exploit the K-means algorithm to generate (Formula presented.) clusters for the colors in the original color table. In the following, our proposed method, respectively, reorders the colors of these clusters by a greedy strategy and concatenates them into the reordered color table. Based on the relationship between the original and the reordered color tables, a novel index matrix can be reconstructed. Finally, state-of-the-art reversible data hiding algorithms can be implemented on the reconstructed index matrix for performance improvement. Since our proposed method improves the reordering effect, enhances the correlation of the reconstructed index matrix, and reduces the length of the encoded location map, the maximal embedding capacities and the visual quality under the fixed embedding capacities are improved. We conducted experiments on two image datasets and six standard images to verify that the performance improvement of our proposed reordering method is better than that of the state-of-the-art methods. © 2025 The Author(s). IET Image Processing published by John Wiley & Sons Ltd on behalf of The Institution of engineering and Technology.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Low-loss Ge-As-S glass fiber for high-power Er:YAG laser transmission and ablation

引用

Optics Express 2025年第10期33卷 20370-20382页

作者： Liang, Xiaolin Wang, Yuyang Jiao, Kai Wang, Xiange Xu, Weisheng Bai, Shengchuang Zhao, Zheming Wang, Rongping Li, Pingxue Shiryaev, Vladimir Wang, Xunsi Laboratory of Infrared Materials and Devices Advanced Technology Research Institute Ningbo University Ningbo315211 China Zhejiang Key Laboratory of Photoelectric Materials and Devices Ningbo315211 China Faculty of Electrical Engineering and Computer Science Ningbo University Ningbo315211 China College of Information and Intelligence Engineering Zhejiang Wanli University Ningbo315000 China College of Data Science Jiaxing University 314001 China Faculty of Materials and Manufacturing Beijing University of Technology Beijing100124 China Devyatykh Institute of Chemistry of High-Purity Substances Russian Academy of Sciences 49 Tropinin Str. Nizhny Novgorod603951 Russia

Ge-based chalcogenide glass fibers exhibit a high damage threshold and exceptional water stability, making them an ideal material for transmitting Er: YAG laser beams. However, the problem of particle scattering in the current glass poses formidable challenge to the transmission of high-power Er: YAG lasers. In this study, we introduce what we believe to be a novel approach to mitigate internal scattering by employing a micron-scale filter in combination with a multi-stage dynamic distillation process, effectively extending the purity limits of Ge-CHG fibers. The loss of Ge-As-S glass has achieved a record of 0.1 dB/m by reducing the internal particle scattering. In addition, the Ge-As-S optical fiber prepared by the isolation extrusion method also exhibits excellent performance in the transmission of Er: YAG laser. The Ge-As-S fiber rod with a core diameter of approximately 750 µm can sustain a laser power of 13.43 W at wavelength of 2.94 µm, i.e., energy density of 684 J/cm2 under atmosphere surrounding. To enhance flexibility, the fiber rod was tapered to gradually reduce its diameter, achieving a core diameter of 300 µm at the narrow end. The flexible Ge-As-S fiber achieved a maximum power output of 4.1 W with an energy density of 580 J/cm2. To our knowledge, this is the first instance of a chalcogenide fiber handling Er: YAG laser power exceeding 5 W flexibly. Finally, we demonstrated the capability of the flexible optical fiber to transmit Er: YAG laser energy for biological tissue ablation, achieving a cutting depth of 3 mm. These results highlight the potential of Ge-As-S fibers as a promising platform for high-power mid-infrared laser transmission, with significant applications in minimally invasive surgery procedures. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.

关键词： Laser surgery

来源：评论

学校读者我要写书评

暂无评论

A Traffic Sign Recognition qnd Speech-BasedFeedback Generation System to Assist Drivers

A Traffic Sign Recognition qnd Speech-BasedFeedback Generati...

引用

Cybernetics, Cognition and Machine Learning Applications (ICCCMLA), IEEE International Conference on

作者： Himangshu Sarma Navanath Saharia Department of Computer Science & Engineering IIIT Sri City Andhra Pradesh Data Engineering Lab CSE IIIT Senapati Manipur

According to the World Health Organization (WHO), 1.3 million people loss their life annually as a result of road accidents. By alerting or assisting the driver while they are driving, information technology such as computer vision can play a significant part in reducing the accident rate. Keeping that in mind, in this article we presents one traffic sign recognition tool using deep neural network (DNN) models with speech-based feedback for vehicle drivers. The proposed system can give voice-based feedback to driver to direct or inform them to take the appropriate action while driving, such as “Drive slow, speed breaker ahead’, “Maximum Speed Limit is 80km/hr“, “School Ahead, drive slow”, “Don't take overtake” etc.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Analysis and Comparison of Image-Based UAV Detection and Identification 9

Analysis and Comparison of Image-Based UAV Detection and Ide...

引用

9th IEEE Uttar Pradesh Section International Conference on Electrical, Electronics and computer engineering, UPCON 2022

作者： Dubey, Nidhish Sai Nithin, Nanduri Mahathi Tripathi, Shrivishal Data Science & Artificial Intelligence Dr. SPM IIIT Naya Raipur India Faculty of Electrical and Computer Engineering Dr. SPM IIIT Naya Raipur India

ISBN: (数字)9798350332506

ISBN: (纸本)9798350332506

This paper presents Unmanned Aerial Vehicles (UAVs) detection and classification with the help of different image-based machine learning modalities. The field of UAVs attracted researchers in recent years in response to the exponential rise in the number of UAVs available in the market with applications ranging from entertainment to defense operations and the risk associated risk by the same. Presently, visual, radar, radio frequency, and acoustic sensing systems are the prevailing technologies in the field of detection and identification of UAVs. The general results of this study show that UAV machine learning-based classifications are propitious and that there are many successful individual contributions. In this research, UAVs were detected and classified using classification methods like Support Vector Machines (SVM), k-nearest neighbor (KNN), and Convolutional Neural Networks (CNN). The results demonstrated that CNN, SVM, and KNN had an accuracy of 91%, 87%, and 78%, respectively. The classifier CNN outperformed other classifiers under the same empirical circumstances. © 2022 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

AI in Medicine: CNN Techniques for Pneumonia and Pandemic Pathogen Identification

AI in Medicine: CNN Techniques for Pneumonia and Pandemic Pa...

引用

Advanced Computing & Communication Technologies (ICACCTech), International Conference on

作者： Ganji Ramanjaiah Nagendar Yamsani Smritilekha Das Balajee Maram Ananda Babu Rudrubati Katakam Venkateswara Rao Dept. of Computer Science & Engineering (Data Science) R VR & JC College of Engineering Chowdavaram Guntur Andhra Pradesh India School of Computer Science and Artificial Intelligence SR University Warangal Telangana Department of CSE Koneru Lakshmaiah Education Foundation Vaddeswaram AP India Dept. Of Computer Science & Engineering (AI & DS) Kallam Haranadhareddy Institute of Technology Chowdavaram Guntur Andhra Pradesh India Dept of Computer Science and Engineering Koneru lakshmaiah Education Foundation Vaddeswaram Guntur DT Andhra Pradesh India

ISBN: (数字)9798331519056

ISBN: (纸本)9798331519063

The current paper deals with the applicability of Convolutional Neural Networks in the detection of pneumonia and pandemic pathogens using medical imaging data. Leveraging the transformative power of AI in our study, we have placed a focus on the development and evaluation of a CNN model that will be trained on a heterogeneous dataset of chest X-rays and CT scans. This paper will propose a method for developing a CNN to classify images from a diverse dataset of chest X-rays and CT scans, ensuring diagnostic accuracy and speed to meet the critical challenges involved in disease diagnosis in a timely manner. Our results show that the CNN model has high accuracy, sensitivity, and specificity for the identification of pneumonia and differentiation of pandemic pathogens from others, surpassing conventional diagnostic techniques. What was more important to underline in this study was that the model demonstrated strong performance across multiple datasets and, more importantly, can easily be transplanted to clinical workflows. Improvement in the early detection capacity made this research emblematic of the huge role AI is expected to play in redefining health care, especially with respiratory diseases and pandemic preparedness. Future work will extend to the augmentation of the dataset and the fine-tuning of the model for its full clinical application.

关键词： Pathogens Pneumonia Accuracy Pandemics Computed tomography Sensitivity and specificity Convolutional neural networks Artificial intelligence Medical diagnostic imaging X-ray imaging

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-guided Mutual Consistency Training for Semi-supervised Biomedical Relation Extraction

Uncertainty-guided Mutual Consistency Training for Semi-supe...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Mao, Bing Jia, Chang Huang, Yucheng He, Kai Wu, Jialun Gong, Tieliang Li, Chen Xi'an Jiaotong University School of Computer Science and Technology Xi'an China Xi'an Jiaotong University Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi'an China

ISBN: (纸本)9781665468190

Biomedical relation extraction seeks to automatically extract biomedical relations from biomedical text, which plays an important role in biomedical studies. However, constructing high-quality biomedical annotation data is not only time-consuming but also requires a high level of knowledge in the biomedical field. To alleviate this problem, Semi-supervised Biomedical Relation Extraction aims to extract relation facts from the limited labeled data and the more readily available unlabeled samples. Existing works can be roughly categorized as self-training methods and self-ensembling methods. The former aims to generate pseudo labels, which may lead to the gradual drift problem. The latter aims to encourage the output of one model to be consistent with the other model, where the acquisition of the model is tedious. To alleviate these issues, we propose a novel Uncertainty-Guided Mutual Consistency Training framework(UG-MCT) for semi-supervised Biomedical relation extraction. Specifically, our framework consists of two models with the same structure, which differ only when updating their weights, and then an intersecting pseudo-label mechanism is designed to convert the prediction discrepancies of the two models into mutual consistency training loss, thus promoting the consistency of model predictions. In addition, we utilize uncertainty as guided information to assist the model in focusing on the confident pseudo labels and mitigate the noise of inaccurate pseudo labeling during training. Thus, our model is very simple and efficient while mitigating the noise introduced by pseudo-labels. UG-MCT is evaluated on multiple datasets in different settings and the experimental results demonstrate that our method is highly effective in semi-supervised biomedical relation extraction compared to the state-of-the-art. © 2022 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Gradual domain adaptation via manifold-constrained distributionally robust optimization 24

Gradual domain adaptation via manifold-constrained distribut...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Amirhossein Saberi Amir Najafi Amin Behjati Ala Emrani Yasaman Zolfi Mahdi Shadrooy Abolfazl Motahari Babak H. Khalaj Department of Electrical Engineering and Sharif Center for Information Systems and Data Science Sharif University of Technology Tehran Iran Department of Computer Engineering Sharif University of Technology Tehran Iran

ISBN: (纸本)9798331314385

The aim of this paper is to address the challenge of gradual domain adaptation within a class of manifold-constrained data distributions. In particular, we consider a sequence of T ≥ 2 data distributions P1, ..., PT undergoing a gradual shift, where each pair of consecutive measures Pi, Pi+1 are close to each other in Wasserstein distance. We have a supervised dataset of size n sampled from P0, while for the subsequent distributions in the sequence, only unlabeled i.i.d. samples are available. Moreover, we assume that all distributions exhibit a known favorable attribute, such as (but not limited to) having intra-class soft/hard margins. In this context, we propose a methodology rooted in Distributionally Robust Optimization (DRO) with an adaptive Wasserstein radius. We theoretically show that this method guarantees the classification error across all Pis can be suitably bounded. Our bounds rely on a newly introduced compatibility measure, which fully characterizes the error propagation dynamics along the sequence. Specifically, for inadequately constrained distributions, the error can exponentially escalate as we progress through the gradual shifts. Conversely, for appropriately constrained distributions, the error can be demonstrated to be linear or even entirely eradicated. We have substantiated our theoretical findings through several experimental results.

关键词：

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：