检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Zhan, Jingtao Xie, Xiaohui Mao, Jiaxin Liu, Yiqun Guo, Jiafeng Zhang, Min Ma, Shaoping Department of Computer Science and Technology Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China Beijing Key Laboratory of Big Data Management and Analysis Methods Gaoling School of Artificial Intelligence Renmin University of China Beijing100872 China CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China

A retrieval model should not only interpolate the training data but also extrapolate well to the queries that are different from the training data. While neural retrieval models have demonstrated impressive performance on ad-hoc search benchmarks, we still know little about how they perform in terms of interpolation and extrapolation. In this paper, we demonstrate the importance of separately evaluating the two capabilities of neural retrieval models. Firstly, we examine existing ad-hoc search benchmarks from the two perspectives. We investigate the distribution of training and test data and find a considerable overlap in query entities, query intent, and relevance labels. This finding implies that the evaluation on these test sets is biased toward interpolation and cannot accurately reflect the extrapolation capacity. Secondly, we propose a novel evaluation protocol to separately evaluate the interpolation and extrapolation performance on existing benchmark datasets. It resamples the training and test data based on query similarity and utilizes the resampled dataset for training and evaluation. Finally, we leverage the proposed evaluation protocol to comprehensively revisit a number of widely-adopted neural retrieval models. Results show models perform differently when moving from interpolation to extrapolation. For example, representation-based retrieval models perform almost as well as interaction-based retrieval models in terms of interpolation but not extrapolation. Therefore, it is necessary to separately evaluate both interpolation and extrapolation performance and the proposed resampling method serves as a simple yet effective evaluation tool for future IR studies. Copyright © 2022, The Authors. All rights reserved.

关键词： Extrapolation

来源：评论

学校读者我要写书评

暂无评论

CubeFlow: Money laundering detection with coupled tensors

arXiv

引用

arXiv 2021年

作者： Sun, Xiaobing Zhang, Jiabao Zhao, Qiming Liu, Shenghua Chen, Jinglei Zhuang, Ruoyu Shen, Huawei Cheng, Xueqi CAS Key Laboratory of Network Data Science and Technology Insititute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences Beijing China China Construction Bank Fintech China

Money laundering (ML) is the behavior to conceal the source of money achieved by illegitimate activities, and always be a fast process involving frequent and chained transactions. How can we detect ML and fraudulent activity in large scale attributed transaction data (i.e. tensors)? Most existing methods detect dense blocks in a graph or a tensor, which do not consider the fact that money are frequently transferred through middle accounts. CubeFlow proposed in this paper is a scalable, flow-based approach to spot fraud from a mass of transactions by modeling them as two coupled tensors and applying a novel multi-attribute metric which can reveal the transfer chains accurately. Extensive experiments show CubeFlow outperforms state-of-the-art baselines in ML behavior detection in both synthetic and real data. Copyright © 2021, The Authors. All rights reserved.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

MVAM: Multi-View Attention Method for Fine-grained Image-Text Matching

arXiv

引用

arXiv 2024年

作者： Cui, Wanqing Cheng, Rui Guo, Jiafeng Cheng, Xueqi CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Alibaba Group Beijing China

Existing two-stream models, such as CLIP, encode images and text through independent representations, showing good performance while ensuring retrieval speed, have attracted attention from industry and academia. However, the single representation often struggles to capture complex content fully. Such models may ignore fine-grained information during matching, resulting in suboptimal retrieval results. To overcome this limitation and enhance the performance of two-stream models, we propose a Multi-View Atention Method (MVAM) for image-text matching. This approach leverages diverse attention heads with unique view codes to learn multiple representations for images and text, which are then concatenated for matching. We also incorporate a diversity objective to explicitly encourage attention heads to focus on distinct aspects of the input data, capturing complementary fine-grained details. This diversity enables the model to represent image-text pairs from multiple perspectives, ensuring a more comprehensive understanding and alignment of critical content. Our method allows models to encode images and text from different perspectives and focus on more critical details, leading to better matching performance. Our experiments on MSCOCO and Flickr30K demonstrate enhancements over existing models, and further case studies reveal that different attention heads can focus on distinct content, achieving more comprehensive representations. © 2024, CC BY.

关键词： Image representation

来源：评论

学校读者我要写书评

暂无评论

Are Neural Ranking Models Robust?

arXiv

引用

arXiv 2021年

作者： Wu, Chen Zhang, Ruqing Guo, Jiafeng Fan, Yixing Cheng, Xueqi CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing NO. 6 Kexueyuan South Road Haidian District Beijing100190 China

Recently, we have witnessed the bloom of neural ranking models in the information retrieval (IR) field. So far, much effort has been devoted to developing effective neural ranking models that can generalize well on new data. There has been less attention paid to the robustness perspective. Unlike the effectiveness which is about the average performance of a system under normal purpose, robustness cares more about the system performance in the worst case or under malicious operations instead. When a new technique enters into the real-world application, it is critical to know not only how it works in average, but also how would it behave in abnormal situations. So we raise the question in this work: Are neural ranking models robust? To answer this question, firstly, we need to clarify what we refer to when we talk about the robustness of ranking models in IR. We show that robustness is actually a multi-dimensional concept and there are three ways to define it in IR: 1) The performance variance under the independent and identically distributed (I.I.D.) setting;2) The out-of-distribution (OOD) generalizability;and 3) The defensive ability against adversarial operations. The latter two definitions can be further specified into two different perspectives respectively, leading to 5 robustness tasks in total. Based on this taxonomy, we build corresponding benchmark datasets, design empirical experiments, and systematically analyze the robustness of several representative neural ranking models against traditional probabilistic ranking models and learning-to-rank (LTR) models. The empirical results show that there is no simple answer to our question. While neural ranking models are less robust against other IR models in most cases, some of them can still win 2 out of 5 tasks. This is the first comprehensive study on the robustness of neural ranking models. We believe the way we study the robustness as well as our findings would be beneficial to the IR community. We will also

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Optimal algorithm for profiling dynamic arrays with finite values

arXiv

引用

arXiv 2018年

作者： Yang, Dingcheng Yu, Wenjian Deng, Junhui Liu, Shenghua BNRist Dept. Computer Science & Tech. Tsinghua Univ. Beijing China CAS Key Lab. Network Data Science & Tech. Inst. Computing Technology Chinese Academy of Sciences Beijing China

How can one quickly answer the most and top popular objects at any time, given a large log stream in a system of billions of users? It is equivalent to find the mode and top-frequent elements in a dynamic array corresponding to the log stream. However, most existing work either restrain the dynamic array within a sliding window, or do not take advantages of only one element can be added or removed in a log stream. Therefore, we propose a profiling algorithm, named S-Profile, which is of O(1) time complexity for every updating of the dynamic array, and optimal in terms of computational complexity. With the profiling results, answering the queries on the statistics of dynamic array becomes trivial and fast. With the experiments of various settings of dynamic arrays, our accurate S-Profile algorithm outperforms the well-known methods, showing at least 2X speedup to the heap based approach and 13X or larger speedup to the balanced tree based approach. Copyright © 2018, The Authors. All rights reserved.

关键词： Computational complexity

来源：评论

学校读者我要写书评

暂无评论

How medical crowdfunding helps people? A large-scale case study on waterdrop fundraising

arXiv

引用

arXiv 2020年

作者： Huang, Junjie Shen, Huawei Cao, Qi Cai, Li Cheng, Xueqi CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Beijing Zongqing Xiangqian Science and Technology Co Ltd Beijing China

While online medical crowdfunding achieved tremendous success, quantitative study about whether and how medical crowdfunding helps people remains little explored. In this paper, we empirically study how online medical crowdfunding helps people using more than 27, 000 fundraising cases in Waterdrop Fundraising, one of the most popular online medical crowdfunding platforms in China. We find that the amount of money obtained by fundraisers is broadly distributed, i.e., a majority of lowly donated cases coexist with a handful of very successful cases. We further investigate the factors that potentially correlate with the success of medical fundraising cases. Profile information of fundraising cases, e.g., geographic information of fundraisers, affects the donated amounts, since detailed description may increase the credibility of a fundraising case. One prominent finding lies in the effect of social network on the success of fundraising cases: the spread of fundraising information along social network is a key factor of fundraising success, and the social capital of fundraisers play an important role in fundraising. Finally, we conduct prediction of donations using machine learning models, verifying the effect of potential factors to the success of medical crowdfunding. Altogether, this work presents a data-driven view of medical fundraising on the web and opens a door to understanding medical crowdfunding. Copyright © 2020, The Authors. All rights reserved.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

TransUKAN:Computing-Efficient Hybrid KAN-Transformer for Enhanced Medical Image Segmentation

arXiv

引用

arXiv 2024年

作者： Wu, Yanlin Li, Tao Wang, Zhihong Kang, Hong He, Along Tianjin Key Laboratory of Network and Data Security Technology College of Computer Science Nankai University Tianjin China Haihe Lab of ITAI China

U-Net is currently the most widely used architecture for medical image segmentation. Benefiting from its unique encoder-decoder architecture and skip connections, it can effectively extract features from input images to segment target regions. The commonly used U-Net is typically based on convolutional operations or Transformers, modeling the dependencies between local or global information to accomplish medical image analysis tasks. However, convolutional layers, fully connected layers, and attention mechanisms used in this process introduce a significant number of parameters, often requiring the stacking of network layers to model complex nonlinear relationships, which can impact the training process. To address these issues, we propose TransUKAN. Specifically, we have improved the KAN to reduce memory usage and computational load. On this basis, we explored an effective combination of KAN, Transformer, and U-Net structures. This approach enhances the model’s capability to capture nonlinear relationships by introducing only a small number of additional parameters and compensates for the Transformer structure’s deficiency in local information extraction. We validated TransUKAN on multiple medical image segmentation tasks. Experimental results demonstrate that TransUKAN achieves excellent performance with significantly reduced parameters. The code will be available athttps://***/wuyanlin-wyl/TransUKAN. Copyright © 2024, The Authors. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Quantum supremacy circuit simulation on sunway taihulight

arXiv

引用

arXiv 2018年

作者： Li, Riling Wu, Bujiao Ying, Mingsheng Sun, Xiaoming Yang, Guangwen Department of Computer Science & Technology Tsinghua University Beijing China CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China Centre for Quantum Software and Information University of Technology Sydney Sydney Australia State Key Lab of Computer Science Institute of Software Chinese Academy of Sciences Department of Computer Science & Technology Tsinghua University Beijing China

With the rapid progress made by industry and academia, quantum computers with dozens of qubits or even larger size are being realized. However, the fidelity of existing quantum computers often sharply decreases as the circuit depth increases. Thus, an ideal quantum circuit simulator on classical computers, especially on high-performance computers, is needed for benchmarking and validation. We design a large-scale simulator of universal random quantum circuits, often called "quantum supremacy circuits", and implement it on Sunway TaihuLight. The simulator can be used to accomplish the following two tasks: 1) Computing a complete output state-vector;2) Calculating one or a few amplitudes. We target the simulation of 49-qubit circuits. For task 1), we successfully simulate such a circuit of depth 39, and for task 2) we reach the 55-depth level. To the best of our knowledge, both of the simulation results reach the largest depth for 49-qubit quantum supremacy circuits. Copyright © 2018, The Authors. All rights reserved.

关键词： Circuit simulation

来源：评论

学校读者我要写书评

暂无评论

Priority-Based Optimization of I/O Isolation for Hybrid Deployed Services 15th

Priority-Based Optimization of I/O Isolation for Hybrid Depl...

引用

15th EAI International Conference on Collaborative Computing: networking, Applications and Worksharing, CollaborateCom 2019

作者： Zhang, Jiancheng Li, Youhuizi Zhou, Li Ren, Zujie Wan, Jian Wang, Yuan Key Laboratory of Complex Systems Modeling and Simulation Ministry of Education Hangzhou Dianzi University Hangzhou China School of Computer Science and Technology Hangzhou Dianzi University Hangzhou China Zhejiang Lab Hangzhou China School of Information and Electronic Engineering Zhejiang University of Science and Technology Hangzhou China Key Enterprise Research Institute of NetEase Big Data of Zhejiang Province Netease Hangzhou Network Co.Ltd Hangzhou China

ISBN: (纸本)9783030301453

With the increasing of software complexity and user demands, collaborative service is becoming more and more popular. Each service focuses on its own specialty, their cooperation can support complicated task with high efficiency. To improve the resources utilization, virtualization technology like container is used and it enables multiple services running in the same physical machine. However, since the host physical machine is shared by several services, the resource competition is inevitable. Isolation is an effective solution, but the weak isolation mechanisms of container cannot handle such complicated scenarios. In the worst situation, the performance of services cannot meet the requirements and the system may crash. In order to solve this problem, we propose a priority-based optimization mechanism for I/O isolation after analyzing the characteristics of typical service workloads. Based on the real-time performance data, priority is automatically assigned to each service and corresponding optimization methods are applied. We evaluate the optimization effects of the priority-based mechanism in both static and dynamic workload cases, besides, the influence of different priority order is also analyzed. The experimental results show that our approach can indeed improve the system performance and guarantee the requirements of all the running services are satisfied. © 2019, ICST Institute for Computer sciences, Social Informatics and Telecommunications Engineering.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Large Language Models are Good Attackers: Efficient and Stealthy Textual Backdoor Attacks

arXiv

引用

arXiv 2024年

作者： Li, Ziqiang Zeng, Yueqi Xia, Pengfei Liu, Lei Fu, Zhangjie Li, Bin Nanjing University of Information Science and Technology China Big Data and Decision Lab University of Science and Technology of China China CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System China

With the burgeoning advancements in the field of natural language processing (NLP), the demand for training data has increased significantly. To save costs, it has become common for users and businesses to outsource the labor-intensive task of data collection to third-party entities. Unfortunately, recent research has unveiled the inherent risk associated with this practice, particularly in exposing NLP systems to potential backdoor attacks. Specifically, these attacks enable malicious control over the behavior of a trained model by poisoning a small portion of the training data. Unlike backdoor attacks in computer vision, textual backdoor attacks impose stringent requirements for attack stealthiness. However, existing attack methods meet significant trade-off between effectiveness and stealthiness, largely due to the high information entropy inherent in textual data. In this paper, we introduce the Efficient and Stealthy Textual backdoor attack method, EST-Bad, leveraging Large Language Models (LLMs). Our EST-Bad encompasses three core strategies: optimizing the inherent flaw of models as the trigger, stealthily injecting triggers with LLMs, and meticulously selecting the most impactful samples for backdoor injection. Through the integration of these techniques, EST-Bad demonstrates an efficient achievement of competitive attack performance while maintaining superior stealthiness compared to prior methods across various text classifier datasets. © 2024, CC BY.

关键词： Deep neural networks

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：