检索结果-内蒙古大学图书馆

Proceedings of the 2025 3rd International Conference on Communication Networks and Machine Learning

作者： Xinpeng Lei Fucheng Wan Key Laboratory of Linguistic and Cultural Computing Ministry of Education Northwest Minzu University Lanzhou Gansu China Key Laboratory of Minzu Languages and Cultures Intelligent Information Processing Northwest Minzu University Lanzhou Gansu China

ISBN: (纸本)9798400713231

The continuous development of large language models has led to tremendous performance improvement and has made them increasingly more important. More and more researchers are starting to explore the use of these large language models in developing artificial intelligence agents. The goal of this research is to develop a question-answering system based on a multi-agent setup with the open-source large language model Qwen2.5. In this multi-agent setup, the two agents include a student and a teacher where the student solves issues with guidance from the teacher. We evaluate accuracy difference between a single-agent and a multi-agent setup with the Chinese evaluation data, C-Eval. This is a departure from other approaches like fine-tuning large models or utilizing methods like chain of thought (CoT) and n-shot learning that are tailor-made to improve model performance. Rather, it is a more general approach aimed at further testing the abilities of large language models and thereby improve their logical reasoning capabilities. The experimental results demonstrate a significant improvement in accuracy in multi-agent systems compared to single-agent setup. In particular, improvement in accuracy over various subjects is around 2.5%. In addition, it has been found that improvement is directly proportional to model parameter size; for example, a model with a parameter size of 14 billion has a 22.18% improvement in accuracy for hard cases and reflects a significant improvement in performance.

关键词： Multi-agent

来源：评论

学校读者我要写书评

暂无评论

FoodSky: A food-oriented large language model that can pass the chef and dietetic examinations

引用

Patterns 2025年第5期6卷

作者： Zhou, Pengfei Min, Weiqing Fu, Chaoran Jin, Ying Huang, Mingyu Li, Xiangyang Mei, Shuhuan Jiang, Shuqiang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Institute of Intelligent Computing Technology Chinese Academy of Sciences Suzhou China Zhongke Shenjian Technology Co. Ltd. Suzhou China

Food is the cornerstone of both survival and social life. With the increasing complexity of global dietary cultures, there is a growing demand for food intelligence to enable tasks like recipe recommendations and diet-disease correlation discovery. To address this, we introduce the food-oriented large language model (LLM) FoodSky, which offers fine-grained perception and reasoning on food data. We constructed a food corpus, FoodEarth, from various authoritative sources to enhance FoodSky's knowledge. We also developed the topic-based selective state space model and hierarchical topic retrieval augmented generation algorithms to improve FoodSky's ability to capture fine-grained food semantics and generate context-aware food-relevant text. Extensive experiments show that FoodSky significantly outperforms general-purpose LLMs on the Chinese National Chef Examination and Dietetic Examination, achieving an accuracy of 83.3% and 91.2%, respectively. Beyond enhancing culinary creativity and promoting healthier eating patterns, FoodSky aims to establish a new benchmark for domain-specific LLMs in addressing real-world food-related challenges. © 2025 The Author(s)

关键词： Food ingredients

来源：评论

学校读者我要写书评

暂无评论

CLIMS++: Cross Language Image Matching with Automatic Context Discovery for Weakly Supervised Semantic Segmentation

引用

International Journal of Computer Vision 2025年 1-20页

作者： Xie, Jinheng Deng, Songhe Hou, Xianxu Luo, Zhaochuan Shen, Linlin Huang, Yawen Zheng, Yefeng Shou, Mike Zheng Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Computer Science Wenzhou-Kean University Wenzhou325060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Show Lab National University of Singapore Singapore639798 Singapore School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215400 China Jarvis Research Center Tencent YouTu Lab Shenzhen518703 China

While promising results have been achieved in weakly-supervised semantic segmentation (WSSS), limited supervision from image-level tags inevitably induces discriminative reliance and spurious relations between target classes and background regions. Thus, Class Activation Map (CAM) usually tends to activate discriminative object regions and falsely includes lots of class-related backgrounds. Without pixel-level supervisions, it could be very difficult to enlarge the foreground activation and suppress those false activation of background regions. In this paper, we propose a novel framework of Cross Language Image Matching with Automatic Context Discovery (CLIMS++), based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress class-related background regions in CAM. In particular, we design object, background region, and text label matching losses to guide the model to excite more reasonable object regions of each category. In addition, we propose to automatically find spurious relations between foreground categories and backgrounds, through which a background suppression loss is designed to suppress the activation of class-related backgrounds. The above designs enable the proposed CLIMS++ to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 datasets show that our CLIMS++ significantly outperforms the previous state-of-the-art methods. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

High-order diversity feature learning for pedestrian attribute recognition

引用

Neural Networks 2025年 188卷 107463页

作者： Wu, Junyi Huang, Yan Gao, Min Niu, Yuzhen Chen, Yuzhong Wu, Qiang Key Laboratory of Network Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou University Fujian Fuzhou China Australian Artificial Intelligence Institute University of Technology Sydney NSW Australia Fujian Key Lab for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering Fuzhou University Fujian Fuzhou China School of Electrical and Data Engineering University of Technology Sydney NSW Australia

Pedestrian attribute recognition (PAR) involves accurately identifying multiple attributes present in pedestrian images. There are two main approaches for PAR: part-based method and attention-based method. The former relies on existing segmentation or region detection methods to localize body parts and learn corresponding attribute-specific feature from the corresponding regions, where the performance heavily depends on the accuracy of body region localization. The latter adopts the embedded attention modules or transformer attention to exploit detailed feature. However, it can focus on certain body regions but often provide coarse attention, failing to capture fine-grained details, the learned feature may also be interfered with by irrelevant information. Meanwhile, these methods overlook the global contextual information. This work argues for replacing coarse attention with detailed attention and integrating it with global contextual feature from ViT to jointly represent attribute-specific regions. To tackle this issue, we propose a High-order Diversity Feature Learning (HDFL) method for PAR based on ViT. We utilize a polynomial predictor to design an Attribute-specific Detailed Feature Exploration (ADFE) module, which can construct the high-order statistics and gain more fine-grained feature. Our ADFE module is a parameter-friendly method that provides flexibility in deciding its utilization during the inference phase. A Soft-redundancy Perception Loss (SPLoss) is proposed to adaptively measure the redundancy between feature of different orders, which can promote diverse characterization of features. Experiments on several PAR datasets show that our method achieves a new state-of-the-art (SOTA) performance. On the most challenging PA100K dataset, our method outperforms previous SOTA by 1.69% and achieves the highest mA of 84.92%. © 2025 The Authors

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

THREE-PRECISION ITERATIVE REFINEMENT WITH PARAMETER REGULARIZATION AND PREDICTION FOR SOLVING LARGE SPARSE LINEAR SYSTEMS

arXiv

引用

arXiv 2025年

作者： Ge, Jifeng Zhang, Juan Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Hunan KeyLaboratory for Computation and Simulation in Science and Engineering School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

This study presents a novel mixed-precision iterative refinement algorithm, GADI-IR, within the general alternating-direction implicit (GADI) framework, designed for efficiently solving large-scale sparse linear systems. By employing low-precision arithmetic, particularly half-precision (FP16), for computationally intensive inner iterations, the method achieves substantial acceleration while maintaining high numerical accuracy. key challenges such as overflow in half-precision and convergence issues for low precision are addressed through careful backward error analysis and the application of a regularization parameter α. Furthermore, the integration of low-precision arithmetic into the parameter prediction process, using Gaussian Process Regression (GPR), significantly reduces computational time without degrading performance. The method is particularly effective for large-scale linear systems arising from discretized partial differential equations and other high-dimensional problems, where both accuracy and efficiency are critical. Numerical experiments demonstrate that the use of mixed-precision strategies not only accelerates computation but also ensures robust convergence, making the approach advantageous for various applications. The results highlight the potential of leveraging lower-precision arithmetic to achieve superior computational efficiency in high-performance computing. © 2025, CC BY.

关键词： Convergence of numerical methods

来源：评论

学校读者我要写书评

暂无评论

Agent-SiMT: Agent-Assisted Simultaneous Translation With Large Language Models

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language processing 2025年 33卷 2074-2083页

作者： Shoutao Guo Shaolei Zhang Zhengrui Ma Min Zhang Yang Feng Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences (ICT/CAS) Beijing China University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology Soochow University Suzhou China Key Laboratory of AI Safety Chinese Academy of Sciences Beijing China

Simultaneous Machine Translation (SiMT) generates target translations in real-time while reading the source sentence. It relies on a policy to determine the optimal timing for producing translations, aiming to achieve both low response latency and high translation quality. Existing SiMT methods typically utilize the traditional Transformer architecture, employing complicated dynamic programming algorithms to jointly optimize policy decisions and translation generation. While these methods excel at determining policies, their translation capabilities are suboptimal. In contrast, Large Language Models (LLMs), trained on extensive corpora, exhibit exceptional generation capabilities but struggle to learn translation policies through traditional Transformer-based training methods. To address these limitations, we introduce Agent-SiMT, a novel framework combining the strengths of LLMs and traditional SiMT methods. Agent-SiMT contains a policy-decision agent and a translation agent. The policy-decision agent is managed by a SiMT model, which determines the policy using partial source sentences and translations. The translation agent, powered by an LLM, generates the target translation according to the policy decisions. These two agents work collaboratively to perform SiMT. Experiments demonstrate that our method attains state-of-the-art performance.

关键词： Translation Hidden Markov models Transformers Training Large language models Collaboration Vocabulary Speech processing Memory modules Machine translation

来源：评论

学校读者我要写书评

暂无评论

Learning with Open-world Noisy Data via Class-independent Margin in Dual Representation Space

arXiv

引用

arXiv 2025年

作者： Pan, Linchao Gao, Can Zhou, Jie Wang, Jinbao College of Computer Science and Software Engineering Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China

Learning with Noisy Labels (LNL) aims to improve the model generalization when facing data with noisy labels, and existing methods generally assume that noisy labels come from known classes, called closed-set noise. However, in real-world scenarios, noisy labels from similar unknown classes, i.e., open-set noise, may occur during the training and inference stage. Such open-world noisy labels may significantly impact the performance of LNL methods. In this study, we propose a novel dual-space joint learning method to robustly handle the open-world noise. To mitigate model overfitting on closed-set and open-set noises, a dual representation space is constructed by two networks. One is a projection network that learns shared representations in the prototype space, while the other is a One-Vs-All (OVA) network that makes predictions using unique semantic representations in the class-independent space. Then, bi-level contrastive learning and consistency regularization are introduced in two spaces to enhance the detection capability for data with unknown classes. To benefit from the memorization effects across different types of samples, class-independent margin criteria are designed for sample identification, which selects clean samples, weights closed-set noise, and filters open-set noise effectively. Extensive experiments demonstrate that our method outperforms the state-of-the-art methods and achieves an average accuracy improvement of 4.55% and an AUROC improvement of 6.17% on CIFAR80N. Copyright © 2025, The Authors. 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

Big-Moe: Bypassing Isolated Gating For Generalized Multimodal Face Anti-Spoofing

Big-Moe: Bypassing Isolated Gating For Generalized Multimoda...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Yingjie Ma Zitong Yu Xun Lin Weicheng Xie Linlin Shen College of Computer Science and Software Engineering Shenzhen University Great Bay University National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Guangdong Provincial Key Laboratory of Intelligent Information Processing

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In the domain of facial recognition security, multimodal Face Anti-Spoofing (FAS) is essential for countering presentation attacks. However, existing technologies encounter challenges due to modality biases and imbalances, as well as domain shifts. Our research introduces a Mixture of Experts (MoE) model to address these issues effectively. We identified three limitations in traditional MoE approaches to multimodal FAS: (1) Coarse-grained experts’ inability to capture nuanced spoofing indicators; (2) Gated networks’ susceptibility to input noise affecting decision-making; (3) MoE’s sensitivity to prompt tokens leading to overfitting with conventional learning methods. To mitigate these, we propose the Bypass Isolated Gating MoE (BIG-MoE) framework, featuring: (1) Fine-grained experts for enhanced detection of subtle spoofing cues; (2) An isolation gating mechanism to counteract input noise; (3) A novel differential convolutional prompt bypass enriching the gating network with critical local features, thereby improving perceptual capabilities. Extensive experiments on four benchmark datasets demonstrate significant generalization performance improvement in multimodal FAS task. The code is released at https://***/murInJ/BIG-MoE.

关键词： Convolutional codes Sensitivity Convolution Face recognition Noise Logic gates Feature extraction Security Speech processing Overfitting

来源：评论

学校读者我要写书评

暂无评论

Digital Human Techniques for Education Reform 24

Digital Human Techniques for Education Reform

引用

7th International Conference on Educational Technology Management, ICETM 2024

作者： Guo, Peirong Zhang, Qi Tian, Chunwei Xue, Wanli Feng, Xiaocheng School of Software Northwestern Polytechnical University Shaanxi Xi’an China School of Economics and Management Harbin Institute of Technology at Weihai Shandong Weihai China Shenzhen Research Institute of Northwestern Polytechnical University Northwestern Polytechnical University Guangdong Shenzhen China School of Computer Science and Engineering Tianjin University of Technology Tianjin China Research Center for Social Computing and Information Retrieval Harbin Institute of Technology Heilongjiang Harbin China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen Key Laboratory of Media Security Shenzhen University Guangdong Shenzhen China

ISBN: (纸本)9798400717468

The rapid evolution of artificial intelligence, big data, and generative AI models has ushered in significant transformations across various sectors, including education. Digital Human Technique, an innovative technology grounded in advanced computer science and artificial intelligence, is reshaping educational paradigms by enabling virtual humans to simulate human behavior, express emotions, and interact with users. This paper explores the application of Digital Human Technique in education reform, focusing on creating immersive, intelligent classroom experiences that foster meaningful interactions between teachers and students. We define Digital Human Technique and delve into its key technical components such as character modeling and rendering, natural language processing, computer vision, and augmented reality technologies. Our methodology involves analyzing the role of educational digital humans created through these technologies, assessing their impact on educational processes, and examining various application scenarios in educational reform. Results indicate that Digital Human Technique significantly enhances the learning experience by enabling personalized teaching, increasing engagement, and fostering emotional connections. Educational digital humans serve as virtual teachers, interactive learning aids, and facilitators of emotional interaction, effectively addressing the challenges of traditional educational methods. They also promote a deeper understanding of complex concepts through simulated environments and interactive digital content. © 2024 Copyright held by the owner/author(s).

关键词： Digital elevation model

来源：评论

学校读者我要写书评

暂无评论

Virtualizing Next-Generation Programmable Networks: Techniques, Use Cases, and Promising Future Directions

引用

IEEE Network 2025年第3期39卷 148-155页

作者： Chen, Xiang Du, Hongyang Liu, Wenjing Liu, Hongyan Zhang, Jianshan Yang, Xu Zhang, Dong Wu, Chunming Liu, Xuan Niyato, Dusit Zhejiang University College of Computer Science and Technology Hangzhou310027 China The University of Hong Kong Department of Electrical and Electronic Engineering Pokfulam Hong Kong Zhejiang Media Group Hangzhou310005 China Minjiang University Fujian Provincial Key Laboratory of Information Processing and Intelligent Control College of Computer and Data Science Fuzhou350108 China Fuzhou University College of Computer and Data Science Fuzhou350108 China Yangzhou University College of Information Engineering College of Artificial Intelligence Yangzhou225009 China Nanyang Technological University College of Computing and Data Science Jurong West 639798 Singapore

For decades, network virtualization plays a crucial role in modern networks, e.g., 6G mobile networks: tenants construct arbitrary virtual networks on the same physical network. However, production networks are becoming programmable and need more ways of network configurations. So prior techniques like SDN hypervisors are deprecated since they cannot configure programmable networks. In response, a new category of network virtualization techniques has emerged recently to virtualize programmable networks, i.e., programmable network hypervisors (PNHs). This article first reviews this trend to comprehensively survey PNHs. To understand their benefits, we build a testbed with hardware switches and sophisticated topologies. Our testbed experiments stress-test hypervisors and demonstrate that they outperform prior works with higher acceptance ratios of virtual networks under large-scale scenarios and heavy loads. Finally, we highlight the open issues of these techniques to shed light on the future directions in this promising topic. © 1986-2012 IEEE.

关键词： Virtualization

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：