检索结果-内蒙古大学图书馆

32nd ACM International Conference on Multimedia, MM 2024

作者： Guo, Qian Liang, Xinyan Qian, Yuhua Cui, Zhihua Wen, Jie Shanxi Key Laboratory of Big Data Analysis and Parallel Computing Taiyuan University of Science and Technology Taiyuan China Institute of Big Data Science and Industry Shanxi University Taiyuan China

ISBN: (纸本)9798400706868

In multi-modal classification tasks, a good fusion algorithm can effectively integrate and process multi-modal data, thereby significantly improving its performance. Researchers often focus on the design of complex fusion operators and have proposed numerous fusion operators, while paying less attention to the design of feature fusion usage, specifically how features should be fused to better facilitate multi-modal classification tasks. In this article, we propose a progressive skip reasoning fusion network (PSRFN) to make some attempts to address this issue. Firstly, unlike most existing multi-modal fusion methods that only use one fusion operator in a single stage to fuse all view features, PSRFN utilizes the progressive skip reasoning (PSR) block to fuse all views with a fusion operator at each layer. Specifically, each PSR block utilizes all view features and the fused features from the previous layer to jointly obtain the fused features for the current layer. Secondly, each PSR block utilizes a dual-weighted fusion strategy with learnable parameters to adaptively allocate weights during the fusion process. The first level of weighting assigns weights to each view feature, while the second level assigns weights to the fused features from the previous layer and the fused features obtained from the first level of weighting in the current layer. This strategy ensures that the PSR block can dynamically adjust the weights based on the actual contribution of features. Finally, to enable the model to fully utilize feature information from different levels for feature fusion, the skip connections are adopted between PSR blocks. Extensive experiment results on six real multi-modal datasets show that a better usage for fusion operator is indeed able to improve performance. © 2024 ACM.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

InstAttention: In-Storage Attention Offloading for Cost-Effective Long-Context LLM Inference 31

InstAttention: In-Storage Attention Offloading for Cost-Effe...

引用

31st IEEE International Symposium on High Performance Computer Architecture, HPCA 2025

作者： Pan, Xiurui Li, Endian Li, Qiao Liang, Shengwen Shan, Yizhou Zhou, Ke Luo, Yingwei Wang, Xiaolin Zhang, Jie Peking University China University of Electronic Science and Technology of China China Institute of Computing Technology Chinese Academy of Sciences China Huawei Cloud China Wuhan National Laboratory for Optoelectronics Huazhong University of Science and Technology China

ISBN: (纸本)9798331506476

The widespread of Large Language Models (LLMs) marks a significant milestone in generative AI. Nevertheless, the increasing context length and batch size in offline LLM inference escalate the memory requirement of the key-value (KV) cache, which imposes a huge burden on the GPU VRAM, especially for resource-constrained scenarios (e.g., edge computing). Several cost-effective solutions leverage host memory or SSDs to reduce storage costs for offline inference scenarios and improve the throughput. Nevertheless, they suffer from significant performance penalties imposed by intensive KV cache accesses due to limited PCIe bandwidth. To address these issues, we propose InstAttention, a novel LLM inference system that offloads the most performance-critical computation (i.e., attention in decoding phase) and data (i.e., KV cache) parts to Computational Storage Drives (CSDs), which minimize the enormous KV transfer overheads. InstAttention designs a dedicated flashaware in-storage attention engine with KV cache management mechanisms to exploit the high internal bandwidths of CSDs instead of being limited by the PCIe bandwidth. The optimized P2P transmission between GPU and CSDs further reduces data migration overheads. Experimental results demonstrate that for a 13B model using an NVIDIA A6000 GPU, InstAttention improves throughput for long-sequence inference by up to 11.1 ×, compared to existing SSD-based solutions such as FlexGen. © 2025 IEEE.

关键词： Computer graphics equipment

来源：评论

学校读者我要写书评

暂无评论

Ten Challenging Problems in Federated Foundation Models

引用

IEEE Transactions on Knowledge and Data Engineering 2025年第07期37卷 4314-4337页

作者： Fan, Tao Gu, Hanlin Cao, Xuemei Chan, Chee Seng Chen, Qian Chen, Yiqiang Feng, Yihui Gu, Yang Geng, Jiaxiang Luo, Bing Liu, Shuoling Ong, Win Kent Ren, Chao Shao, Jiaqi Sun, Chuan Tang, Xiaoli Tae, Hong Xi Tong, Yongxin Wei, Shuyue Wu, Fan Xi, Wei Xu, Mingcong Yang, He Yang, Xin Yan, Jiangpeng Yu, Hao Yu, Han Zhang, Teng Zhang, Yifei Zhang, Xiaojin Zheng, Zhenzhe Fan, Lixin WeBank Shenzhen China Southwestern University of Finance and Economic China Universiti Malaya Malaysia Chinese Academy of Sciences Institute of Computing Technology China Duke Kunshan University Kunshan China Institute of Innovation E Fund Management Co. Ltd. China KTH Royal Institute of Technology Sweden Nanyang Technological University Singapore Beihang University China Shanghai Jiao Tong University China Xi'an Jiaotong University China Huazhong University of Science and Technology China Hong Kong University of Science and Technology Department of Computer Science and Engineering Hong Kong

Federated Foundation Models (FedFMs) represent a distributed learning paradigm that fuses general competences of foundation models as well as privacy-preserving capabilities of federated learning. This combination allows the large foundation models and the small local domain models at the remote clients to learn from each other in a teacher-student learning setting. This paper provides a comprehensive summary of the ten challenging problems inherent in FedFMs, encompassing foundational theory, utilization of private data, continual learning, unlearning, Non-IID and graph data, bidirectional knowledge transfer, incentive mechanism design, game mechanism design, model watermarking, and efficiency. The ten challenging problems manifest in five pivotal aspects: "Foundational Theory," which aims to establish a coherent and unifying theoretical framework for FedFMs. "Data," addressing the difficulties in leveraging domain-specific knowledge from private data while maintaining privacy;"Heterogeneity," examining variations in data, model, and computational resources across clients;"Security and Privacy," focusing on defenses against malicious attacks and model theft;and "Efficiency," highlighting the need for improvements in training, communication, and parameter efficiency. For each problem, we offer a clear mathematical definition on the objective function, analyze existing methods, and discuss the key challenges and potential solutions. This in-depth exploration aims to advance the theoretical foundations of FedFMs, guide practical implementations, and inspire future research to overcome these obstacles, thereby enabling the robust, efficient, and privacy-preserving FedFMs in various real-world applications. © 1989-2012 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

A Diffuse-Interface Lattice Boltzmann Method for the Dendritic Growth with Thermosolutal Convection

引用

Communications in Computational Physics 2023年第4期33卷 1164-1188页

作者： Chengjie Zhan Zhenhua Chai Baochang Shi Ping Jiang Shaoning Geng Dongke Sun School of Mathematics and Statistics Huazhong University of Science and TechnologyWuhan430074P.R.China Institute of Interdisciplinary Research for Mathematics and Applied Science Huazhong University of Science and TechnologyWuhan 430074P.R.China Hubei Key Laboratory of Engineering Modeling and Scientific Computing Huazhong University of Science and TechnologyWuhan430074P.R.China The State Key Laboratory of Digital Manufacturing Equipment and Technology School of Mechanical Science and EngineeringHuazhong University of Science and TechnologyWuhan 430074P.R.China School of Mechanical Engineering Southeast UniversityNanjing 211189P.R.China

In this work,we proposed a diffuse-interface model for the dendritic growth with thermosolutal *** this model,the sharp boundary between the fluid and solid dendrite is firstly replaced by a thin but nonzero thickness diffuse interface,which is described by the order parameter,and the diffuse-interface based governing equations for the dendritic growth are *** solve the model for the dendritic growth with thermosolutal convection,we also developed a diffuse-interface multirelaxation-time lattice Boltzmann(LB)*** this method,the order parameter in the phase-field equation is combined into the force caused by the fluid-solid interaction,and the treatment on the complex fluid-solid interface can be *** addition,four LB models are designed for the phase-field equation,concentration equation,temperature equation and the Navier-Stokes equations in a unified ***,we performed some simulations of the dendritic growth to test the present diffuse-interface LB method,and found that the numerical results are in good agreements with some previous works.

关键词： Dendritic growth diffuse interface lattice Boltzmann method phase-field method.

来源：评论

学校读者我要写书评

暂无评论

Physics-informed neural networks with residual/gradient-based adaptive sampling methods for solving partial differential equations with sharp solutions

引用

Applied Mathematics and Mechanics(English Edition) 2023年第7期44卷 1069-1084页

作者： Zhiping MAO Xuhui MENG School of Mathematical Sciences Fujian Provincial Key Laboratory of Mathematical Modeling and High-Performance Scientific ComputingXiamen UniversityXiamen 361005Fujian ProvinceChina Institute of Interdisciplinary Research for Mathematics and Applied Science School of Mathematics and StatisticsHuazhong University of Science and TechnologyWuhan 430074China

We consider solving the forward and inverse partial differential equations(PDEs)which have sharp solutions with physics-informed neural networks(PINNs)in this *** particular,to better capture the sharpness of the solution,we propose the adaptive sampling methods(ASMs)based on the residual and the gradient of the *** first present a residual only-based ASM denoted by ASMⅠ.In this approach,we first train the neural network using a small number of residual points and divide the computational domain into a certain number of sub-domains,then we add new residual points in the sub-domain which has the largest mean absolute value of the residual,and those points which have the largest absolute values of the residual in this sub-domain as new residual *** further develop a second type of ASM(denoted by ASMⅡ)based on both the residual and the gradient of the solution due to the fact that only the residual may not be able to efficiently capture the sharpness of the *** procedure of ASMⅡis almost the same as that of ASMⅠ,and we add new residual points which have not only large residuals but also large *** demonstrate the effectiveness of the present methods,we use both ASMⅠand ASMⅡto solve a number of PDEs,including the Burger equation,the compressible Euler equation,the Poisson equation over an Lshape domain as well as the high-dimensional Poisson *** has been shown from the numerical results that the sharp solutions can be well approximated by using either ASMⅠor ASMⅡ,and both methods deliver much more accurate solutions than the original PINNs with the same number of residual ***,the ASMⅡalgorithm has better performance in terms of accuracy,efficiency,and stability compared with the ASMⅠ*** means that the gradient of the solution improves the stability and efficiency of the adaptive sampling procedure as well as the accuracy of the ***,we also employ the similar adaptive sampling technique fo

关键词： physics-informed neural network(PINN) adaptive sampling high-dimension L-shape Poisson equation accuracy

来源：评论

学校读者我要写书评

暂无评论

Towards Cross-Domain Few-Shot Graph Anomaly Detection 24

Towards Cross-Domain Few-Shot Graph Anomaly Detection

引用

24th IEEE International Conference on Data Mining, ICDM 2024

作者： Chen, Jiazhen Fu, Sichao Zhang, Zhibin Ma, Zheng Feng, Mingbin Wirjanto, Tony S. Peng, Qinmu University of Waterloo Department of Statistic and Actuarial Science Canada School of Electronic Information and Communications Huazhong University of Science and Technology China Institute of Computing Technology Chinese Academy of Sciences China Cheriton School of Computer Science University of Waterloo Canada

ISBN: (纸本)9798331506681

Few-shot graph anomaly detection (GAD) has recently garnered increasing attention, which aims to discern anomalous patterns among abundant unlabeled test nodes under the guidance of a limited number of labeled training nodes. Existing few-shot GAD approaches typically adopt meta-training methods trained on richly labeled auxiliary networks to facilitate rapid adaptation to target networks that possess sparse labels. However, these proposed methods often assume that the auxiliary and target networks exist in the same data distributions-an assumption rarely holds in practical settings. This paper explores a more prevalent and complex scenario of cross-domain few-shot GAD, where the goal is to identify anomalies within sparsely labeled target graphs using auxiliary graphs from a related, yet distinct domain. The challenge here is nontrivial owing to inherent data distribution discrepancies between the source and target domains, compounded by the uncertainties of sparse labeling in the target domain. In this paper, we propose a simple and effective framework, termed CDFS-GAD, specifically designed to tackle the aforementioned challenges. CDFS-GAD first introduces a domain-adaptive graph contrastive learning module, which is aimed at enhancing cross-domain feature alignment. Then, a prompt tuning module is further designed to extract domain-specific features tailored to each domain. Moreover, a domain-adaptive hypersphere classification loss is proposed to enhance the discrimination between normal and anomalous instances under minimal supervision, utilizing domain-sensitive norms. Lastly, a self-training strategy is introduced to further refine the predicted scores, enhancing its reliability in few-shot settings. Extensive experiments on twelve real-world cross-domain data pairs demonstrate the effectiveness of the proposed CDFS-GAD framework in comparison to various existing GAD methods including unsupervised, semi-supervised, few-shot and cross-domain GAD methods. © 20

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Model Trip: Enhancing Privacy and Fairness in Model Fusion Across Multi-Federations for Trustworthy Global Healthcare 40

Model Trip: Enhancing Privacy and Fairness in Model Fusion A...

引用

40th IEEE International Conference on Data Engineering, ICDE 2024

作者： Chen, Qian Chen, Yiqiang Yan, Bingjie Jiang, Xinlong Zhang, Xiaojin Kang, Yan Zhang, Teng Huang, Wuliang Gao, Chenlong Fan, Lixin Yang, Qiang Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Beijing Key Laboratory of Mobile Computing and Pervasive Device China Peng Cheng Laboratory China School of Computer Science and Technology Huazhong University of Science and Technology China Webank Hong Kong Hong Kong University of Science and Technology Hong Kong

ISBN: (纸本)9798350317152

Federated Learning has emerged as a revolutionary innovation in the evolving landscape of global healthcare, fostering collaboration among institutions and facilitating collaborative data analysis. As practical applications continue to proliferate, numerous federations have formed in different regions. The optimization and sustainable development of federation-pretrained models have emerged as new challenges. These challenges primarily encompass privacy, population shift and data dependency, which may lead to severe consequences such as the leakage of sensitive information within models and training samples, unfair model performance and resource burdens. To tackle these issues, we propose FairFusion, a cross-federation model fusion approach that enhances privacy and fairness. FairFusion operates across federations within a Model Trip paradigm, integrating knowledge from diverse federations to continually enhance model performance. Through federated model fusion, multi-objective quantification and optimization, FairFusion obtains trustworthy solutions that excel in utility, privacy and fairness. We conduct comprehensive experiments on three public real-world healthcare datasets. The results demonstrate that FairFusion achieves outstanding model fusion performance in terms of utility and fairness across various model structures and subgroups with sensitive attributes while guaranteeing model privacy. © 2024 IEEE.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

YOLO-RSOD: Improved YOLO Remote Sensing Object Detection 27th

YOLO-RSOD: Improved YOLO Remote Sensing Object Detection

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Xu, Yang Lu, Jun College of Computer Science and Technology Heilongjiang University Harbin150080 China Jiaxiang Industrial Technology Research Institute of Heilongjiang University Shandong Jining272400 China Key Laboratory of Database and Parallel Computing of Heilongjiang Province Heilongjiang University Harbin150080 China

ISBN: (纸本)9783031784460

Remote sensing object detection has important application value in fields such as environmental monitoring and resource detection and analysis. However, the current universal object detectors are not very effective in detecting remote sensing objects. To this end, this paper proposes an efficient, low-complexity and anchor-free remote sensing object detection framework YOLO-RSOD based on YOLOv7. First, an additional Tiny Object Head is proposed for better detection of micro-remote sensing objects. The original Head is then replaced with Decoupled Head (DH) to explore the detection potential of the decoupled detection head structure. Then the Explicit Vision Center (EVC) in the Centralized Feature Pyramid Network (CFP) is added to further improve the detection ability of remote sensing objects. Finally, this article also integrates a global attention module (GAM) to find attention areas in dense object scenes. Ablation experiments on the general remote sensing target detection dataset VisDrone2021 demonstrate the effectiveness of several modules introduced in this paper in remote sensing target detection. On the VisDrone2021 data set, YOLO-RSOD can achieve accuracy rates of 30.7% AP50:95 and 51.7% AP50, which are 3.1% and 3.2% higher than the baseline model respectively. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Object tracking

来源：评论

学校读者我要写书评

暂无评论

False Positive Detection for Text-Based Person Retrieval 21st

False Positive Detection for Text-Based Person Retrieval

引用

21st Pacific Rim International Conference on Artificial Intelligence, PRICAI 2024

作者： Cao, Yan Lu, Jun College of Computer Science and Technology Heilongjiang University Harbin150080 China Jiaxiang Industrial Technology Research Institute of Heilongjiang University Shandong Jining272400 China Key Laboratory of Database and Parallel Computing of Heilongjiang Province Heilongjiang University Harbin150080 China

ISBN: (纸本)9789819601189

Text-based person retrieval (TBPR) is a challenging topic in cross-modal retrieval tasks, aiming to query corresponding person images based on textual descriptions. This task is complicated by noisy correspondences between images and text due to incorrect text annotations and low-quality images. Although many robust noise learning methods have been proposed with satisfactory performance, they tend to directly filter out a portion of the noise in the dataset, which decreases the ability to distinguish the noise as the number of training epochs increases. This paper proposes a noise-robust learning strategy based on false positive detection (FPD). Precisely, FPD consists of two main components: 1) The key-value library (KVL) module utilizes a set of clean samples, assigns confidence weights to each training sample during the training process, and adaptively adjusts the contribution of each sample to reduce the accumulation of noise. 2) To improve the performance of TBPR, this paper designs a local token-based selection method, which utilizes the contribution of each local token to the weights in the self-attention matrix and selects the local tokens with vital information as global representations. Extensive experiments are conducted on three datasets (CUHK-PEDES, ICFG-PEDES, and RSTPReID) to evaluate the performance and robustness of FPD under noisy conditions. The experimental results show that the FPD has better results at high noise ratios, and all the metrics outperform the state-of-the-art method under the RSTPReID dataset with a noise ratio of 80%, where Rank-1 outperforms the best method by 2.28%, 1.64% for mAP, and 1.92% for mINP. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Image annotation

来源：评论

学校读者我要写书评

暂无评论

DCI-Net: Remote Sensing Image-Based Object Detector 27th

DCI-Net: Remote Sensing Image-Based Object Detector

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Cui, Quanyue Lu, Jun College of Computer Science and Technology Heilongjiang University Harbin150080 China Jiaxiang Industrial Technology Research Institute of Heilongjiang University Shandong Jining272400 China Key Laboratory of Database and Parallel Computing of Heilongjiang Province Heilongjiang University Harbin150080 China

ISBN: (纸本)9783031781124

Recently the analysis of remotely sensed images has played a vital role in various aspects of research. The current researches ignore the unique prior knowledge in remote sensing images and do not consider exploring the contextual information of the object, while the existence of multi-scale and high image resolution of objects in remote sensing images also affects the accuracy of the object detection task. Based on the above problems, this paper proposes a object detector DCI-Net (Dynamic Context-Aware IoU Network) based on remote sensing images, in which the proposed CASK (Context-Aware Selective Kernel) module can explicitly model the interdependence between the convolutional feature channels. A loss function Pi_IoU is proposed, which adaptively adjusts the penalty factor in combination with the size of the detected object. A DySample module is introduced, which is able to effectively extract and utilize the spatial structure features. The model in this paper improves the detection accuracy of complex objects in remote sensing images. On the DIOR dataset, compared with the baseline model YOLOV9, the accuracy is improved by 0.6%, the number of parameters is decreased by 4%, and the floating point operation speed is improved by 27.8%. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Optical remote sensing

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：