检索结果-内蒙古大学图书馆

HGFormer: Topology-Aware Vision Transformer with HyperGraph Learning

IEEE Transactions on Multimedia 2025年

作者： Wang, Hao Zhang, Shuo Leng, Biao Beihang University School of Computer Science and Engineering Beijing100191 China Beijing Key Laboratory of Traffic Data Analysis and Mining School of Computer Science & Technology Beijing100044 China

The computer vision community has witnessed an extensive exploration of vision transformers in the past two years. Drawing inspiration from traditional schemes, numerous works focus on introducing vision-specific inductive biases. However, the implicit modeling of permutation invariance and fully-connected interaction with individual tokens disrupts the regional context and spatial topology, further hindering higher-order modeling. This deviates from the principle of perceptual organization that emphasizes the local groups and overall topology of visual elements. Thus, we introduce the concept of hypergraph for perceptual exploration. Specifically, we propose a topology-aware vision transformer called HyperGraph Transformer (HGFormer). Firstly, we present a Center Sampling K-Nearest Neighbors (CS-KNN) algorithm for semantic guidance during hypergraph construction. Secondly, we present a topology-aware HyperGraph Attention (HGA) mechanism that integrates hypergraph topology as perceptual indications to guide the aggregation of global and unbiased information during hypergraph messaging. Using HGFormer as visual backbone, we develop an effective and unitive representation, achieving distinct and detailed scene depictions. Empirical experiments show that the proposed HGFormer achieves competitive performance compared to the recent SoTA counterparts on various visual benchmarks. Extensive ablation and visualization studies provide comprehensive explanations of our ideas and contributions. © 2025 IEEE.

关键词： Hypergraph Learning Image Classification Vision Transformer

来源：评论

学校读者我要写书评

暂无评论

Adaptive Group Personalization for Federated Mutual Transfer Learning 41

Adaptive Group Personalization for Federated Mutual Transfer...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Xu, Haoqing Shen, Dian Wang, Meng Wang, Beilun School of Computer Science and Engineering Southeast University Nanjing210096 China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education China College of Design and Innovation Tongji University Shanghai China

Mutual transfer learning aims to improve prediction with knowledge from related domains. Recently, federated learning is applied in this field to address the communication and privacy concerns. However, previous clustered federated learning (CFL) solutions lack theoretical guarantee of learnability recovery and require time-consuming hyper-parameter tuning, while centralized mutual transfer learning methods lack adaptability to concept drifts. In this paper, we propose the Adaptive Group Personalization method (AdaGrP) to overcome these challenges. We adaptively decide the recovery threshold with a nonparametric method, adaptive threshold correction, for tuning-free solution with relaxed condition. Theoretical results guarantee the perfect learnability recovery with the corrected threshold. Empirical results show AdaGrP achieves 16.9% average improvement in learnability structure recovery compared with state-of-the-art CFL baselines. Copyright 2024 by the author(s)

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph based Mutual Attention for Machine Reading Comprehension over Anti-Terrorism Corpus

引用

data Intelligence 2023年第3期5卷 685-706页

作者： Feng Gao Jin Hou Jinguang Gu Lihua Zhang School of Computer Science and Technology Wuhan University of Science and TechnologyWuhan 430065Hubei The Key Laboratory of Rich-Media Knowledge Organization and Service of Digital Publishing Content Insitute of Scientic and Technical Information of ChinaBeijing 100038China Wuhan University of Science and Technology Big Data Science and Engineering Research Institute Wuhan 430065Hubei Eastchina Jiaotong University Nanchang 330013Jiangxi

Machine reading comprehension has been a research focus in natural language processing and intelligence ***,there is a lack of models and datasets for the MRC tasks in the anti-terrorism ***,current research lacks the ability to embed accurate background knowledge and provide precise *** address these two problems,this paper first builds a text corpus and testbed that focuses on the anti-terrorism domain in a semi-automatic ***,it proposes a knowledge-based machine reading comprehension model that fuses domain-related triples from a large-scale encyclopedic knowledge base to enhance the semantics of the *** eliminate knowledge noise that could lead to semantic deviation,this paper uses a mixed mutual ttention mechanism among questions,passages,and knowledge triples to select the most relevant triples before embedding their semantics into the *** results indicate that the proposed approach can achieve a 70.70%EM value and an 87.91%F1 score,with a 4.23%and 3.35%improvement over existing methods,respectively.

关键词： Machine reading comprehension Anti-terrorism domain Knowledge embedding Knowledge attention Mutual attention

来源：评论

学校读者我要写书评

暂无评论

DC-Instruct: An Effective Framework for Generative Multi-intent Spoken Language Understanding

DC-Instruct: An Effective Framework for Generative Multi-int...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Xing, Bowen Liao, Lizi Huang, Minlie Tsang, Ivor W. Beijing Key Laboratory of Knowledge Engineering for Materials Science School of Computer and Communication Engineering University of Science and Technology Beijing China Singapore Management University Singapore Group Tsinghua University China CFAR and IHPC Agency for Science Technology and Research Singapore College of Computing and Data Science Nanyang Technological University Singapore

ISBN: (纸本)9798891761643

In the realm of multi-intent spoken language understanding, recent advancements have leveraged the potential of prompt learning frameworks. However, critical gaps exist in these frameworks: the lack of explicit modeling of dual-task dependencies and the oversight of task-specific semantic differences among utterances. To address these shortcomings, we propose DC-Instruct, a novel generative framework based on Dual-task Inter-dependent Instructions (DII) and Supervised Contrastive Instructions (SCI). Specifically, DII guides large language models (LLMs) to generate labels for one task based on the other task's labels, thereby explicitly capturing dual-task inter-dependencies. Moreover, SCI leverages utterance semantics differences by guiding LLMs to determine whether a pair of utterances share the same or similar labels. This can improve LLMs on extracting and discriminating task-specific semantics, thus enhancing their SLU reasoning abilities. Extensive experiments on public benchmark datasets show that DC-Instruct markedly outperforms current generative models and state-of-the-art methods, demonstrating its effectiveness in enhancing dialogue language understanding and reasoning. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Online Streaming Features Causal Discovery Algorithm Based on Partial Rank Correlation

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第1期4卷 197-208页

作者： Yang, Jing Jiang, Liufeng Shen, Anbo Wang, Aiguo Key Laboratory of Knowledge Engineering with Big Data of Ministry of Education Hefei University of Technology Hefei230009 China School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China School of Electronic and Information Engineering Foshan University Foshan528011 China

Aimedat the problem of dynamic causal discovery in the era of artificial intelligence, this article combines partial rank correlation coefficients and streaming features in the field of Bayesian network structure learning and proposes a new online streaming feature causal discovery algorithm based on partial rank correlation named the partial rank casual discovery streaming feature based algorithm. This algorithm is not only suitable for Bayesian causal structure learning in dynamic feature spaces generated by sequential streams of features but can also effectively process multivariate linear Gaussian and nonlinear non-Gaussian data. We present three main contributions. First, for arbitrarily distributed datasets, which can be generated by the additive noise model, we proved that the partial rank correlation coefficient can be used as the criterion for the conditional independence test and explored the distribution of corresponding statistics. Second, the PCSDSF algorithm redefined the relevance based on partial rank statistic prospects and then redefined conditional dependence or independence. This method can significantly reduce the number of conditional independence tests and achieves a good time performance. Finally, theoretical analysis and several experiments proved the reliability of the algorithm. A simulation showed that on average, the PCSDSF algorithm outperforms existing algorithms in terms of both the accuracy and time performance. © 2022 IEEE.

关键词： Additive noise

来源：评论

学校读者我要写书评

暂无评论

High-repetition-rate strong-field terahertz source by optical rectification in DSTMS crystals

引用

High Power Laser science and Engineering 2024年第5期12卷 70-75页

作者： Zhuorui Zheng Kang Wang Hongyang Li Xianze Meng Ye Tian Liwei Song State Key Laboratory of High Field Laser Physics and CAS Center for Excellence in Ultra-intense Laser Science Shanghai Institute of Optics and Fine MechanicsChinese Academy of SciencesShanghaiChina Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of SciencesBeijingChina School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghaiChina School of Physics Science and Engineering Tongji UniversityShanghaiChina

We present the generation of high-repetition-rate strong-field terahertz(THz)pulses from a thin 4-N,N-dimethylamino-4’-N’-methyl-stilbazolium 2,4,6-trimethylbenzenesulfonate(DSTMS)organic crystal pumped by an ytterbium-doped yttrium aluminum garnet *** generated THz pulse energy reaches 932.8 nJ at 1 kHz repetition rate,with a conversion efficiency of 0.19%and a peak electric field of 819 kV/*** a repetition rate of 10 kHz,it is able to maintain a peak electric field of 236 kV/cm and an average THz power of 0.77 *** high-repetition-rate,strong-field THz source provides a convenient tool for the study of THz matter manipulation and THz spectroscopy.

关键词： high repetition rate optical rectification strong-field terahertz source

来源：评论

学校读者我要写书评

暂无评论

引用

17th CCF Conference on computer Supported Cooperative Work and Social Computing, ChineseCSCW 2022

作者： Yu, Cizhou Li, Dongsheng Lu, Tun Jiang, Yichuan School of Computer Science Fudan University Shanghai China Shanghai Key Laboratory of Data Science Fudan University Shanghai China School of Computer Science and Engineering Southeast University Nanjing China

ISBN: (纸本)9789819923557

With the rapid development of the Internet, recommendation algorithms are increasingly influencing consumers’ decision-making. The issue of fairness in recommendation algorithms, especially popularity bias, is also becoming more and more heated. For the research on popularity bias, most existing methods cannot balance long-term interactivity and low cost. Using the ABM simulation method, various variables in the recommender system can be controlled, and the long-term impact of the continuous interaction between the recommender system and the user can be studied, especially the influence of popularity bias. In this paper, we construct and implement a recommendation algorithm simulation framework based on ABM, and two algorithms, Item-Based CF and SVD, are respectively deployed on it to count item popularity distribution and Gini coefficient under multiple rounds of recommendation. The indicators, combined with the visualization results of user interest offset, are used to explore the popularity bias problem of two classic recommendation algorithms under multiple rounds of interaction. We also summarize the existing problems and make an outlook for future improvements. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

GraphTransfer: A Generic Feature Fusion Framework for Collaborative Filtering

arXiv

引用

arXiv 2024年

作者： Xia, Jiafeng Li, Dongsheng Gu, Hansu Lu, Tun Gu, Ning Shanghai Key Laboratory of Data Science Fudan University Shanghai China School of Computer Science Fudan University Shanghai China Microsoft Research Asia Shanghai China *** Seattle United States

Graph Neural Networks (GNNs) have demonstrated effectiveness in collaborative filtering tasks due to their ability to extract powerful structural features. However, combining the graph features extracted from user-item interactions and auxiliary features extracted from user genres and item properties remains a challenge. Currently available fusion methods face two major issues: 1) simple methods such as concatenation and summation are generic, but not accurate in capturing feature relationships;2) task-specific methods like attention mechanisms and meta paths may not be suitable for general feature fusion. To address these challenges, we present GraphTransfer, a simple but universal feature fusion framework for GNN-based collaborative filtering. Our method accurately fuses different types of features by first extracting graph features from the user-item interaction graph and auxiliary features from users and items using GCN. The proposed cross fusion module then effectively bridges the semantic gaps between the interaction scores of different features. Theoretical analysis and experiments on public datasets show that GraphTransfer outperforms other feature fusion methods in CF tasks. Additionally, we demonstrate the universality of our framework via empirical studies in three other scenarios, showing that GraphTransfer leads to significant improvements in the performance of CF algorithms. Copyright © 2024, The Authors. All rights reserved.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Stitching, Fine-tuning, Re-training: A SAM-enabled Framework for Semi-supervised 3D Medical Image Segmentation

引用

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Li, Shumeng Qi, Lei Yu, Qian Huo, Jing Shi, Yinghuan Gao, Yang Nanjing University State Key Laboratory for Novel Software Technology China Nanjing University National Institute of Healthcare Data Science China School of Computer Science and Engineering China Shandong Women's University School of Data and Computer Science China

Segment Anything Model (SAM) fine-tuning has shown remarkable performance in medical image segmentation in a fully supervised manner, but requires precise annotations. To reduce the annotation cost and maintain satisfactory performance, in this work, we leverage the capabilities of SAM for establishing semi-supervised medical image segmentation models. Rethinking the requirements of effectiveness, efficiency, and compatibility, we propose a three-stage framework, i.e., Stitching, Fine-tuning, and Re-training (SFR). The current fine-tuning approaches mostly involve 2D slice-wise fine-tuning that disregards the contextual information between adjacent slices. Our stitching strategy mitigates the mismatch between natural and 3D medical images. The stitched images are then used for fine-tuning SAM, providing robust initialization of pseudo-labels. Afterwards, we train a 3D semi-supervised segmentation model while maintaining the same parameter size as the conventional segmenter such as V-Net. Our SFR framework is plug-and-play, and easily compatible with various popular semi-supervised methods. We also develop an extended framework SFR+ with selective fine-tuning and re-training through confidence estimation. Extensive experiments validate that our SFR and SFR+ achieve significant improvements in both moderate annotation and scarce annotation across five datasets. In particular, SFR framework improves the Dice score of Mean Teacher from 29.68% to 74.40% with only one labeled data of LA dataset. The code is available at https://***/ShumengLI/SFR. © 1982-2012 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Communication-Efficient Distributed Learning with Local Immediate Error Compensation

arXiv

引用

arXiv 2024年

作者： Cheng, Yifei Shen, Li Xu, Linli Qian, Xun Wu, Shiwei Zhou, Yiming Zhang, Tie Tao, Dacheng Chen, Enhong The Anhui Province Key Lab of Big Data Analysis and Application State Key Laboratory of Cognitive Intelligence School of Data Science School of Computer Science University of Science and Technology China The JD Explore Academy China The Shanghai Artificial Intelligence Laboratory China

Gradient compression with error compensation has attracted significant attention with the target of reducing the heavy communication overhead in distributed learning. However, existing compression methods either perform only unidirectional compression in one iteration with higher communication cost, or bidirectional compression with slower convergence rate. In this work, we propose the Local Immediate Error Compensated SGD (LIEC-SGD) optimization algorithm to break the above bottlenecks based on bidirectional compression and carefully designed compensation approaches. Specifically, the bidirectional compression technique is to reduce the communication cost, and the compensation technique compensates the local compression error to the model update immediately while only maintaining the global error variable on the server throughout the iterations to boost its efficacy. Theoretically, we prove that LIEC-SGD is superior to previous works in either the convergence rate or the communication cost, which indicates that LIEC-SGD could inherit the dual advantages from unidirectional compression and bidirectional compression. Finally, experiments of training deep neural networks validate the effectiveness of the proposed LIEC-SGD algorithm. Copyright © 2024, The Authors. All rights reserved.

关键词： Error compensation

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：