检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Chen, Haipeng Wu, Sifan Wang, Zhigang Yin, Yifang Jiao, Yingying Lyu, Yingda Liu, Zhenguang College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China College of Computer Science and Technology Zhejiang Gongshang University China A*STAR Singapore Public Computer Education and Research Center Jilin University China The State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China

Video-based human pose estimation has long been a fundamental yet challenging problem in computer vision. Previous studies focus on spatio-temporal modeling through the enhancement of architecture design and optimization strategies. However, they overlook the causal relationships in the joints, leading to models that may be overly tailored and thus estimate poorly to challenging scenes. Therefore, adequate causal reasoning capability, coupled with good interpretability of model, are both indispensable and prerequisite for achieving reliable results. In this paper, we pioneer a causal perspective on pose estimation and introduce a causal-inspired multitask learning framework, consisting of two stages. In the first stage, we try to endow the model with causal spatio-temporal modeling ability by introducing two self-supervision auxiliary tasks. Specifically, these auxiliary tasks enable the network to infer challenging keypoints based on observed keypoint information, thereby imbuing causal reasoning capabilities into the model and making it robust to challenging scenes. In the second stage, we argue that not all feature tokens contribute equally to pose estimation. Prioritizing causal (keypoint-relevant) tokens is crucial to achieve reliable results, which could improve the interpretability of the model. To this end, we propose a Token Causal Importance Selection module to identify the causal tokens and non-causal tokens (e.g., background and objects). Additionally, non-causal tokens could provide potentially beneficial cues but may be redundant. We further introduce a non-causal tokens clustering module to merge the similar non-causal tokens. Extensive experiments show that our method outperforms state-of-the-art methods on three large-scale benchmark datasets. Copyright © 2025, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

SoAy: A Solution-based LLM API-using Methodology for Academic Information Seeking 25

SoAy: A Solution-based LLM API-using Methodology for Academi...

引用

Proceedings of the 31st ACM SIGKDD Conference on knowledge Discovery and data Mining V.1

作者： Yuanchun Wang Jifan Yu Zijun Yao Jing Zhang Yuyang Xie Shangqing Tu Yiyang Fu Youhe Feng Jinkai Zhang Jingyao Zhang Bowen Huang Yuanyao Li Huihui Yuan Lei Hou Juanzi Li Jie Tang School of Information Renmin University of China Beijing China & Key Laboratory of Data Engineering and Knowledge Engineering MOE Beijing China Institute of Education Tsinghua University Beijing China Department of Computer Science and Technology Tsinghua University Beijing China School of Information Renmin University of China Beijing China & Engineering Research Center of Database and Business Intelligence MOE Beijing China School of Information Renmin University of China Beijing China Zhipu AI Beijing China

ISBN: (纸本)9798400712456

Applying large language models (LLMs) to academic API usage shows promise in reducing researchers' efforts to seek academic information. However, current LLM methods for using APIs struggle with the complex API coupling commonly encountered in academic queries. To address this, we introduce SoAy, a solution-based LLM methodology for academic information seeking. SoAy enables LLMs to generate code for invoking APIs, guided by a pre-constructed API calling sequence referred to as a solution. This solution simplifies the model's understanding of complex API relationships, while the generated code enhances reasoning efficiency. LLMs are aligned with this solution-oriented, code-based reasoning method by automatically enumerating valid API coupling sequences and transforming them into queries and executable *** evaluate SoAy, we introduce SoAyBench, an evaluation benchmark accompanied by SoAyEval, built upon a cloned environment of APIs from AMiner. Experimental results demonstrate a 34.58-75.99% performance improvement compared to state-of-the-art LLM API-based baselines. All datasets, codes, tuned models, and deployed online services are publicly accessible at https://***/RUCKBReasoning/SoAy.

关键词： academic information seeking

来源：评论

学校读者我要写书评

暂无评论

DeepSeek: Paradigm Shifts and Technical Evolution in Large AI Models

引用

IEEE/CAA Journal of Automatica Sinica 2025年第5期12卷 841-858页

作者： Luolin Xiong Haofen Wang Xi Chen Lu Sheng Yun Xiong Jingping Liu Yanghua Xiao Huajun Chen Qing-Long Han Yang Tang Key Laboratory of Smart Manufacturing in Energy Chemical Process Ministry of Education and the Engineering Research Center of Process System Engineering Ministry of Education East China University of Science and Technology Shanghai China College of Design and Innovation Tongji University Shanghai China Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China School of Software Beihang University Beijing China School of Information Science and Engineering East China University of Science and Technology Shanghai China AZFT Joint Laboratory for Knowledge Engine Hangzhou Innovation Center College of Computer Science and Technology Zhejiang University Hangzhou China School of Science Computing and Engineering Technologies Swinburne University of Technology Melbourne VIC Australia

DeepSeek, a Chinese artificial intelligence (AI) startup, has released their V3 and R1 series models, which attracted global attention due to their low cost, high performance, and open-source advantages. This paper begins by reviewing the evolution of large AI models focusing on paradigm shifts, the mainstream large language model (LLM) paradigm, and the DeepSeek paradigm. Subsequently, the paper highlights novel algorithms introduced by DeepSeek, including multi-head latent attention (MLA), mixture-of-experts (MoE), multi-token prediction (MTP), and group relative policy optimization (GRPO). The paper then explores DeepSeek's engineering breakthroughs in LLM scaling, training, inference, and system-level optimization architecture. Moreover, the impact of DeepSeek models on the competitive AI landscape is analyzed, comparing them to mainstream LLMs across various fields. Finally, the paper reflects on the insights gained from DeepSeek's innovations and discusses future trends in the technical and engineering development of large AI models, particularly in data, training, and reasoning.

关键词： Training Technological innovation Foundation models Large language models Reinforcement learning Prediction algorithms Market research Cognition Inference algorithms Optimization

来源：评论

学校读者我要写书评

暂无评论

Poster abstract: data-enriched simulation of data management applications for wireless sensor networks

Poster abstract: Data-enriched simulation of data management...

引用

8th ACM International Conference on Embedded Networked Sensor Systems, SenSys 2010

作者： Liu, Yu Li, Jianzhong Gao, Hong Shi, Shengfei Data and Knowledge Engineering Research Center Harbin Institute of Technology Harbin China

ISBN: (纸本)9781450303446

Simulation is an essential means for evaluating WSN applications. As many WSN applications embrace in-network data processing functionalities, more sophisticated simulation tools with data-enriched test case scenarios, such as extensive environment input and accurate hardware models, are necessary. This work devises TOSSIM DB, a data-enriched WSN simulation framework, allowing unmodified applications to be evaluated as if virtually deployed in a third-party environment. Notable features of TOSSIM DB include i) integrating third-party sensor datasets as virtual test scenarios, ii) emulating the statistical properties of sensor devices, and iii) seamless integration with the TinyOS tool-chain. Copyright 2010 ACM.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

Carbon-doped percentage effect on the mechanical properties of nanoporous silicon sample using molecular dynamics simulation

引用

Case Studies in Chemical and Environmental engineering 2025年 11卷

作者： Sawaran Singh, Narinderjit Singh Ali, Ali B.M. Ameen, Hawzhen Fateh M. Al-Zahiwat, Murtadha M. Salahshour, Soheil Emami, Nafiseh Faculty of Data Science and Information Technology INTI International University Persiaran Perdana BBN Putra Nilai Nilai 71800 Malaysia Air Conditioning Engineering Department College of Engineering University of Warith Al-Anbiyaa Karbala Iraq Department of Petroleum Technology Erbil Technology College Erbil Polytechnic University Kurdistan Region Erbil Iraq Department of Petroleum Engineering Knowledge University Erbil Iraq Department of Chemical engineering College of Engineering University of Misan Amarah Iraq Faculty of Engineering and Natural Sciences Istanbul Okan University Istanbul Türkiye Faculty of Engineering and Natural Sciences Bahcesehir University Istanbul Türkiye Research Center of Applied Mathematics Khazar University Baku Türkiye Fast Computing Center Shabihsazan Ati Pars Tehran Iran

Porous materials have attracted considerable attention from researchers due to its many uses in molecular separation, heterogeneous catalysis, absorption technologies, and electronic improvements. These solid materials, often defined by their structural voids, are essential in several sectors. This research investigated the impact of carbon doping on the mechanical characteristics of nanoporous silicon matrices. The use of high-purity silicon doping is very beneficial in the semiconductor industry and is crucial for high-power devices and automotive applications. This study simulates a nanoporous silicon sample by molecular dynamics methods, adding carbon doping at different concentrations. The findings demonstrate that when the carbon doping concentration escalated from 1 % to 30 %, the mechanical resistance of the system decreased correspondingly. The ultimate tensile strength fell from 10.26 to 9.02 GPa. Furthermore, Young's modulus rose from 83.47 to 98.37 GPa. The decline in mechanical stability was associated with a drop in the model's total weight, which had considerable ramifications for industrial applications. Thus, incorporating C-doped nanoporous silicon into real applications not only lowered the weight of target materials but also improved their use. © 2025 The Author(s)

关键词： Carbon doping Mechanical properties Molecular dynamics simulation Nanoporous Porous materials

来源：评论

学校读者我要写书评

暂无评论

Improving performance by creating a native join-index for OLAP

引用

Frontiers of Computer Science 2011年第2期5卷 236-249页

作者： Yansong ZHANG Shan WANG Jiaheng LU National Survey Research Center at Renmin University of China Beijing 100872 China Key Laboratory of the Ministry of Education for Data Engineering and Knowledge Engineering Renmin University of China Beijing 100872 China School of Information Renmin University of China Beijing 100872 China

The performance of online analytical processing （OLAP） is critical for meeting the increasing requirements of massive volume analytical applications. Typical techniques, such as in-memory processing, column-storage, and join indexes focus on high perfor- mance storage media, efficient storage models, and reduced query processing. While they effectively perform OLAP applications, there is a vital limitation： main- memory database based OLAP （MMOLAP） cannot provide high performance for a large size data set. In this paper, we propose a novel memory dimension table model, in which the primary keys of the dimension table can be directly mapped to dimensional tuple addresses. To achieve higher performance of dimensional tuple access, we optimize our storage model for dimension tables based on OLAP query workload features. We present directly dimensional tuple accessing （DDTA） based join （DDTA- JOIN）, a technique to optimize query processing on the memory dimension table by direct dimensional tuple access. We also contribute by proposing an optimization of the predicate tree to shorten predicate operation length by pruning useless predicate processing. Our experimental results show that the DDTA-JOIN algorithm is superior to both simulated row-store main memory query processing and the open-source column-store main memory database MonetDB, thanks to the reduced join cost and simple yet efficient query processing.

关键词： directly dimensional tuple accessing (DDTA) – DDTA JOIN – native join index – predicate tree

来源：评论

学校读者我要写书评

暂无评论

Conceptual clustering of social bookmarking sites

Conceptual clustering of social bookmarking sites

引用

Workshop on Learning, knowledge, and Adaptivity, LWA 2007

作者： Grahl, Miranda Hotho, Andreas Stumme, Gerd Knowledge and Data Engineering Group University of Kassel Germany Research Center L3S Hannover Germany

ISBN: (纸本)9783860109076

Currently, social bookmarking systems provide intuitive support for browsing locally their content. A global view is usually presented by the tag cloud of the system, but it does not allow a conceptual drill-down, e. g., along a conceptual hierarchy. In this paper, we present a clustering approach for computing such a conceptual hierarchy for a given folksonomy. The hierarchy is complemented with ranked lists of users and resources most related to each cluster. The rankings are computed using our FolkRank algorithm. We have evaluated our approach on large scale data from the *** bookmarking system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

On the predictability of talk attendance at academic conferences (Extended Abstract) 16

On the predictability of talk attendance at academic confere...

引用

16th Workshops on Learning, knowledge, Adaptation, LWA 2014: knowledge Discovery, data Mining and Machine Learning, KDML 2014, Information Retrieval, IR 2014 and knowledge Management, FGWM 2014

作者： Scholz, Christoph Illig, Jens Atzmueller, Martin Stumme, Gerd University of Kassel ITeG Research Center Knowledge and Data Engineering Group Wilhelmshöher Allee 73 Kassel Germany

来源：评论

学校读者我要写书评

暂无评论

Myocardial Ischemia Diagnosis Using a Reduced Lead System 40

Myocardial Ischemia Diagnosis Using a Reduced Lead System

引用

40th Annual International Conference of the IEEE engineering in Medicine and Biology Society, EMBC 2018

作者： Aranda Hernandez, A. Bonizzi, P. Karel, J. Peeters, R. Department of Research and Technology Bakken Research Center Medtronic Maastricht Netherlands Department of Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

ISBN: (纸本)9781538636466

This research presents a novel statistical model for diagnosing acute myocardial infarction (AMI). The model is based on features extracted from a reduced lead system consisting of a subset of three leads from the standard 12-lead ECG. We selected a set of relevant parameters commonly used in the clinical practice for ECG-based AMI diagnosis, namely ST elevation and T-wave maximum. We also selectedfeatures, not used in clinical practice, that were derived from vectorcardiography and computed on the reduced three-lead system (pseudo-VCG parameters). To validate the model, we used 104 patients coming from the Physionet STAFF III database which contains 12-lead ECG recordings at baseline and in coronary artery occlusion condition during angioplasty (PTCA). Results show that pseudo-VCG features are able to diagnose AMI slightly better than ST elevation and T-wave maximum features together (area under the ROC curve (AUC) 0.87 vs AUC 0.85). When combining pseudo-VCG features together with ST elevation, and T-wave maximum, the performance improved significantly (AUC 0.95, sensitivity 89.6% and specificity 82.7%). Results indicate a potential for diagnosing AMI using the proposed reduced lead system and the selected set of features. We suggest its possible use for diagnosing AMI in long-term, ambulatory and home monitoring situations, allowing an earlier and faster diagnosis. © 2018 IEEE.

关键词： Electrocardiography Feature extraction Heart Myocardium databases Lead Standards

来源：评论

学校读者我要写书评

暂无评论

Integrating Physical Prediction Methods and AI-based Satellite data Analysis Methods in Earthquake Damage Estimation 8th

Integrating Physical Prediction Methods and AI-based Satelli...

引用

8th International Symposium on Reliability engineering and Risk Management, ISRERM 2022

作者： Miyamoto, Takashi Department of Civil and Environmental Engineering University of Yamanashi Japan Smart Data and Knowledge Services German Research Center for Artificial Intelligence Germany

ISBN: (纸本)9789811851841

In order to estimate the damage distribution immediately after an earthquake, both physical prediction methods and data-driven methods that analyze sensing data obtained from satellites are used. However, the former has the problem of prediction accuracy, while the latter has the problem of difficulty in detecting detailed damage patterns such as partial destruction. Therefore, we presents a method that improves the detection accuracy of detailed damage distribution of structures such as total and partial collapse by integrating both methods. As an integration scheme of the two methods, a data assimilation method based on Bayes’ theorem is adopted in this study. We proposed a method to update the damage probability of each structure obtained from physical simulations by conditioning it on the observed data obtained from satellite image analysis, and verified its effectiveness. © 2022 ISRERM Organizers. Published by research Publishing, Singapore.

关键词： Prediction models

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：