检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Deng, Kaijun Zheng, Dezhi Xie, Jindong Wang, Jinbao Xie, Weicheng Shen, Linlin Song, Siyang Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China Department of Computer Science University of Exeter United Kingdom

Accurately synthesizing talking face videos and capturing fine facial features for individuals with long hair presents a significant challenge. To tackle these challenges in existing methods, we propose a decomposed per-embedding Gaussian fields (DEGSTalk), a 3D Gaussian Splatting (3DGS)-based talking face synthesis method for generating realistic talking faces with long hairs. Our DEGSTalk employs Deformable Pre-Embedding Gaussian Fields, which dynamically adjust pre-embedding Gaussian primitives using implicit expression coefficients. This enables precise capture of dynamic facial regions and subtle expressions. Additionally, we propose a Dynamic Hair-Preserving Portrait Rendering technique to enhance the realism of long hair motions in the synthesized videos. Results show that DEGSTalk achieves improved realism and synthesis quality compared to existing approaches, particularly in handling complex facial dynamics and hair preservation. Our code will be publicly available at https://***/CVI-SZU/DEGSTalk. Copyright © 2024, The Authors. All rights reserved.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Automated data Visualization from Natural Language via Large Language Models: An Exploratory Study

arXiv

引用

arXiv 2024年

作者： Wu, Yang Wan, Yao Zhang, Hongyu Sui, Yulei Wei, Wucai Zhao, Wei Xu, Guandong Jin, Hai Huazhong University of Science and Technology China Chongqing University China University of New South Wales Australia University of Technology Sydney Australia National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China

The Natural Language to Visualization (NL2Vis) task aims to transform natural-language descriptions into visual representations for a grounded table, enabling users to gain insights from vast amounts of data. Recently, many deep learning-based approaches have been developed for NL2Vis. Despite the considerable efforts made by these approaches, challenges persist in visualizing data sourced from unseen databases or spanning multiple tables. Taking inspiration from the remarkable generation capabilities of Large Language Models (LLMs), this paper conducts an empirical study to evaluate their potential in generating visualizations, and explore the effectiveness of in-context learning prompts for enhancing this task. In particular, we first explore the ways of transforming structured tabular data into sequential text prompts, as to feed them into LLMs and analyze which table content contributes most to the NL2Vis. Our findings suggest that transforming structured tabular data into programs is effective, and it is essential to consider the table schema when formulating prompts. Furthermore, we evaluate two types of LLMs: finetuned models (e.g., T5-Small) and inference-only models (e.g., GPT-3.5), against state-of-the-art methods, using the NL2Vis benchmarks (i.e., nvBench). The experimental results reveal that LLMs outperform baselines, with inference-only models consistently exhibiting performance improvements, at times even surpassing fine-tuned models when provided with certain few-shot demonstrations through in-context learning. Finally, we analyze when the LLMs fail in NL2Vis, and propose to iteratively update the results using strategies such as chain-of-thought, role-playing, and code-interpreter. The experimental results confirm the efficacy of iterative updates and hold great potential for future study. © 2024, CC BY.

关键词： data visualization

来源：评论

学校读者我要写书评

暂无评论

A neural kernel framework for relation extraction

引用

Expert systems with Applications 2025年 286卷

作者： Wang, Kai Chen, Yanping Yang, Weizhe Huang, Ruizhang Qin, Yongbin Text Computing & Cognitive Intelligence Engineering Research Center of National Education Ministry College of Computer Science and Technology Guizhou University China State Key Laboratory of Public Big Data Guizhou University China

Relation extraction (RE) poses significant challenges due to the complexity of identifying semantic relationships between overlapping entity pairs within sentences. Traditional kernel-based methods effectively leverage instance-level similarities but heavily depend on manually designed kernel functions, requiring substantial domain expertise and limiting their flexibility. On the other hand, neural network approaches excel at automatically learning abstract feature representations but typically neglect critical instance-specific information, potentially missing valuable relational details during inference. These limitations motivate the exploration of a hybrid approach that effectively integrates the strengths of traditional kernels with the representational power of neural networks. In this paper, we propose a neural kernel framework designed to bridge this critical gap. The proposed framework employs neural kernels, whose parameters are initialized based on task-specific objectives and optimized through neural training procedures, enabling adaptive learning of similarity measures without relying on handcrafted kernel functions. By maintaining instance-level details through annotated training examples, neural kernels create discriminative, flexible decision boundaries tailored specifically to the relation extraction task. We further develop three complementary neural kernel components, instance kernel, description kernel, and cluster kernel, to show the advantages of kernel substitution. Extensive experiments on the ACE 2005, SemEval 2010, and CoNLL 2004 datasets demonstrate that our neural kernel framework significantly outperforms existing state-of-the-art methods, achieving F1-scores of 88.11 %, 91.08 %, and 98.63 %, respectively. © 2025 Elsevier Ltd

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

GraphFly: Efficient Asynchronous Streaming Graphs Processing via Dependency-Flow

GraphFly: Efficient Asynchronous Streaming Graphs Processing...

引用

Supercomputing Conference

作者： Dan Chen Chuangyi Gui Yi Zhang Hai Jin Long Zheng Yu Huang Xiaofei Liao National Engineering Research Center for Big Data Technology and System/Services Computing Technology and System Lab/Clusters and Grid Computing Lab Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9781665454452

Existing streaming graph processing systems typically adopt two phases of refinement and recomputation to ensure the correctness of the incremental computation. However, severe redundant memory accesses exist due to the unnecessary synchronization among independent edge updates. In this paper, we present GraphFly, a high-performance asynchronous streaming graph processing system based on dependency-flows. GraphFly features three key designs: 1) Dependency trees (D-trees), which helps quickly identify independent graph updates with low cost; 2) Dependency-flow based processing model, which exploits the space-time dependent co-scheduling for cache efficiency; 3) Specialized graph data layout, which further reduces memory accesses. We evaluate GraphFly, and the results show that GraphFly significantly outperforms state-of-the-art systems KickStarter and GraphBolt by 5.81× and 1.78× on average, respectively. Also, GraphFly scales well with different sizes of update batch and compute resources.

关键词： Costs High performance computing Memory management Layout data models Synchronization

来源：评论

学校读者我要写书评

暂无评论

HMvisor: dynamic hybrid memory management for virtual machines

引用

Science China(Information Sciences) 2021年第9期64卷 74-89页

作者： Dang YANG Haikun LIU Hai JIN Yu ZHANG National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing LabSchool of Computer Science and Technology Huazhong University of Science and Technology

Emerging non-volatile memory(NVM) technologies promise high density, low cost and dynamic random access memory(DRAM)-like performance, at the expense of limited write endurance and high write energy consumption. It is more practical to use NVM combining with the traditional DRAM. However, the hybrid memory management such as page migration becomes more challenging in a virtualization environment because virtual machines(VMs) are unaware of the memory heterogeneity. In this paper, we propose HMvisor, a hypervisor and VM coordinated hybrid memory management mechanism to better utilize DRAM and NVM resources. HMvisor exposes the memory heterogeneity to VMs by mapping virtual NUMA nodes to different physical NUMA nodes. We propose a lightweight and efficient page migration mechanism by decoupling page hotness tracking from page migration. HMvisor performs those operations in the hypervisor and VMs separately, without disrupting the execution of VMs. We also propose a memory resource trading policy to adjust the capacity of DRAM and NVM for each VM, with the monetary cost unchanged. We implement our prototype system based on QEMU/KVM and evaluate it with several benchmarks. Experimental results show that HMvisor can reduce 50% of write traffic to NVM with less than 5% performance overhead. Moreover, the hybrid memory adjustment scheme in HMvisor can significantly improve application performance by up to 30×.

关键词： hypervisor virtual machine non-volatile memory hybrid memory management

来源：评论

学校读者我要写书评

暂无评论

Towards Efficient data-Centric Robust Machine Learning with Noise-based Augmentation

arXiv

引用

arXiv 2022年

作者： Liu, Xiaogeng Wang, Haoyu Zhang, Yechao Wu, Fangzhou Hu, Shengshan School Of Cyber Science And Engineering National Engineering Research Center For Big Data Technology And System Services Computing Technology And System Lab Hubei Engineering Research Center On Big Data Security Huazhong University Of Science And Technology Wuhan430074 China

The data-centric machine learning aims to find effective ways to build appropriate datasets which can improve the performance of AI models. In this paper, we mainly focus on designing an efficient data-centric scheme to improve robustness for models towards unforeseen malicious inputs in the black-box test settings. Specifically, we introduce a noisedbased data augmentation method which is composed of Gaussian Noise, Salt-and-Pepper noise, and the PGD adversarial perturbations. The proposed method is built on lightweight algorithms and proved highly effective based on comprehensive evaluations, showing good efficiency on computation cost and robustness enhancement. In addition, we share our insights about the data-centric robust machine learning gained from our experiments. Copyright © 2022, The Authors. All rights reserved.

关键词： Gaussian noise (electronic)

来源：评论

学校读者我要写书评

暂无评论

Layered Structure Aware Dependent Microservice Placement Toward Cost Efficient Edge Clouds

Layered Structure Aware Dependent Microservice Placement Tow...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE Computer and Communications Societies

作者： Deze Zeng Hongmin Geng Lin Gu Zhexiong Li School of Computer Science China University of Geosciences Wuhan China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

Although the containers are featured by light-weightness, it is still resource-consuming to pull and startup a large container image, especially in relatively resource-constrained edge cloud. Fortunately, Docker, as the most widely used container, provides a unique layered architecture that allows the same layer to be shared between microservices so as to lower the deployment cost. Meanwhile, it is highly desirable to deploy dependent microservices of an application together to lower the operation cost. Therefore, the balancing of microservice deployment cost and the operation cost should be considered comprehensively to achieve minimal overall cost of an on-demand application. In this paper, we first formulate this problem into a Quadratic Integer Programming form (QIP) and prove it as a NP-hard problem. We further propose a Randomized Rounding-based Microservice Deployment and Layer Pulling (RR-MDLP) algorithm with low computation complexity and guaranteed approximation ratio. Through extensive experiments, we verify the high efficiency of our algorithm by the fact that it significantly outperforms existing state-of-the-art microservice deployment strategies.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Bidirectional Weighted Co-Association-Based Clustering Ensemble Algorithm for High-Speed Train Delay Prediction 29th

Bidirectional Weighted Co-Association-Based Clustering Ensem...

引用

29th International Conference on Computational and Experimental engineering and Sciences, ICCES 2023

作者： Zhao, Yi Bai, Muchen Ni, Panpan Jiang, Yunan Feng, Xuan Zhang, Xiaobo School of Design SouthWest Jiaotong University Chengdu611756 China School of Computing and Artificial Intelligence SouthWest Jiaotong University Chengdu611756 China Sichuan Institute of Computer Sciences Chengdu610042 China National Engineering Laboratory of Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu611756 China Manufacturing Industry Chains Collaboration and Information Support Technology Key Laboratory of Sichuan Province Southwest Jiaotong University Chengdu611756 China

ISBN: (纸本)9783031425141

In recent years, high-speed trains are one of the popular research topics in the field of transportation and have attracted much attention from scholars at home and abroad in view of their high efficiency and safety features. High-speed trains are susceptible to unexpected situations of delayed arrival due to the disturbance of external factors including wind and temperature during the trip. The delay of high-speed trains has a vital impact on people's work efficiency and living standards. To solve the problem of intelligent forecast of high-speed train delay, we design and implement a clustering ensemble algorithm based on text data of bullet train operation environment monitoring. We firstly combine the concepts of cosine similarity, Jaccard similarity and information entropy, then fuse a bidirectional weighting formula, and further propose a novel Clustering Ensemble algorithm based on Bidirectional Weighted Co-association (CE-BWC). This approach consists of an inter-cluster weighting formula used to obtain the implicit generalized information by weighting each cluster between clusters, and a reliability weighting formula transferring the cluster information to the data item itself to measure the correlation degree of the data group. Experiments prove that our CE-BWC has better prediction performance for high-speed train delay and outperforms the state-of-the-art benchmark algorithms. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Research on Image Captioning Based on Vision-language Pre-trained Models

Research on Image Captioning Based on Vision-language Pre-tr...

引用

International Conference on big data and Information Analytics (bigDIA)

作者： Shiwei Zou Yuxiang Xie Jie Yan Yingmei Wei Xidao Luan Department of System Engineering National University of Defense Technology Hunan China Key Laboratory of Big Data Department of System Engineering National University of Defense Technology Hunan Changsha China Department of Computer Science Changsha University Hunan China

Image captioning is a vision-language task that targets at describing an image by generating a coherent sentence automatically. This technology allows computers to understand and describe images like humans, enabling further processing of images. It combines computer vision and natural language processing, which has broad applications. This paper investigates the use of vision-language pre-trained models for image captioning. In this article, an encoder-decoder hybrid model based on PVT and BERT is pre-trained on a large number of image-text pairs, possessing both understanding and generation capabilities. It consists of three components: 1) an image encoder and a text encoder to extract image and text features respectively; 2) a multi-modal encoder to align the two modalities; and 3) a text decoder to generate text descriptions. This paper focuses on enhancing the image encoder of the model, which is crucial for image captioning models as it extracts high-resolution image features. Reducing training time and computational costs can also improve the applicability of pre-trained models. Improvements are made to the original model by modifying the ViT model architecture of the image encoding module. A feature pyramid structure and a spatial reduction attention mechanism is introduced to extract high-resolution image features while reducing computational complexity and memory usage, thereby enhancing the universality of the model. The model adopts a pure Transformer architecture, discarding the convolutional structure, and achieves excellent performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation

arXiv

引用

arXiv 2024年

作者： He, Weizhao Zhang, Yang Zhuo, Wei Shen, Linlin Yang, Jiaqi Deng, Songhe Sun, Liang Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China School of Computer Science University of Nottingham China

Few-shot semantic segmentation (FSS) endeavors to segment unseen classes with only a few labeled samples. Current FSS methods are commonly built on the assumption that their training and application scenarios share similar domains, and their performances degrade significantly while applied to a distinct domain. To this end, we propose to leverage the cutting-edge foundation model, the Segment Anything Model (SAM), for generalization enhancement. The SAM however performs unsatisfactorily on domains that are distinct from its training data, which primarily comprise natural scene images, and it does not support automatic segmentation of specific semantics due to its interactive prompting mechanism. In our work, we introduce APSeg, a novel auto-prompt network for cross-domain few-shot semantic segmentation (CD-FSS), which is designed to be auto-prompted for guiding cross-domain segmentation. Specifically, we propose a Dual Prototype Anchor Transformation (DPAT) module that fuses pseudo query prototypes extracted based on cycle-consistency with support prototypes, allowing features to be transformed into a more stable domain-agnostic space. Additionally, a Meta Prompt Generator (MPG) module is introduced to automatically generate prompt embeddings, eliminating the need for manual visual prompts. We build an efficient model which can be applied directly to target domains without fine-tuning. Extensive experiments on four cross-domain datasets show that our model outperforms the state-of-the-art CD-FSS method by 5.24% and 3.10% in average accuracy on 1-shot and 5-shot settings, respectively. Copyright © 2024, The Authors. 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

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：