检索结果-内蒙古大学图书馆

CLIMS++: Cross Language Image Matching with Automatic Context Discovery for Weakly Supervised Semantic Segmentation

International Journal of computer Vision 2025年 1-20页

作者： Xie, Jinheng Deng, Songhe Hou, Xianxu Luo, Zhaochuan Shen, Linlin Huang, Yawen Zheng, Yefeng Shou, Mike Zheng Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Computer Science Wenzhou-Kean University Wenzhou325060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Show Lab National University of Singapore Singapore639798 Singapore School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215400 China Jarvis Research Center Tencent YouTu Lab Shenzhen518703 China

While promising results have been achieved in weakly-supervised semantic segmentation (WSSS), limited supervision from image-level tags inevitably induces discriminative reliance and spurious relations between target classes and background regions. Thus, Class Activation Map (CAM) usually tends to activate discriminative object regions and falsely includes lots of class-related backgrounds. Without pixel-level supervisions, it could be very difficult to enlarge the foreground activation and suppress those false activation of background regions. In this paper, we propose a novel framework of Cross Language Image Matching with Automatic Context Discovery (CLIMS++), based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress class-related background regions in CAM. In particular, we design object, background region, and text label matching losses to guide the model to excite more reasonable object regions of each category. In addition, we propose to automatically find spurious relations between foreground categories and backgrounds, through which a background suppression loss is designed to suppress the activation of class-related backgrounds. The above designs enable the proposed CLIMS++ to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 datasets show that our CLIMS++ significantly outperforms the previous state-of-the-art methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Data Integration with XML ETL Processing

Data Integration with XML ETL Processing

引用

2020 International Conference on computer science, Engineering and Applications, ICCSEA 2020

作者： Jayashree, G. Priya, C. Technology and Advanced Studies Department of Computer Science Chennai India VELS Institute of Science Technology and Advanced Studies Department of Information Technology Chennai India

ISBN: (纸本)9781728158303

Data forms the key success enabler for any business organization. Structured collection and organized data along with this analytical enablement will provide more insights to the organization's existing operations and ways for better improvement. In order to do efficient data collection, integration of data from various sources forms the vital role. Many of the vital business initiatives for an organization requires the data to integrate. Such initiatives performed by creating a customized solution across every data source. In such given architecture, the complexity to integrate data source is proportionate to the increase in data sources. Use of eXtensible Markup Language (XML) for standardizing the information exchange across several sources is an industry standard implementation. It is the technology used to implement and deliver an enterprise-level data integration solution using heterogeneous data sources. The transformation capability of XML is capable of handling very complex and big documents and can do processing of ~thousands of real-time transactions using message bus or Web Services methods. The XML provided solutions are flexible and easily adaptable, thus enabling them to incorporate into a broader range of organizational solutions. This paper discusses in detail, the data integration methods and their challenges and processing of XML data using Extraction, Transformation and Load method. The paper also provides the several ways to parse the XML data as part of the data processing with the case study for an On Line Transaction Process (OLTP) system of a manufacturing company. © 2020 IEEE.

关键词： XML

来源：评论

学校读者我要写书评

暂无评论

Explanations of Large Language Models Explain Language Representations in the Brain

arXiv

引用

arXiv 2025年

作者： Rahimi, Maryam Yaghoobzadeh, Yadollah Daliri, Mohammad Reza Biomedical Engineering Department School of Electrical Engineering Iran University of Science and Technology Tehran Iran Electrical and Computer Engineering Department University of Tehran Tehran Iran School of Cognitive Sciences Institute for Research in Fundamental Sciences Tehran Iran Tehran Institute for Advanced Studies Khatam University Tehran Iran

Large language models (LLMs) not only exhibit human-like performance but also share computational principles with the brain’s language processing mechanisms. While prior research has focused on mapping LLMs’ internal representations to neural activity, we propose a novel approach using explainable AI (XAI) to strengthen this link. Applying attribution methods, we quantify the influence of preceding words on LLMs’ next-word predictions and use these explanations to predict fMRI data from participants listening to narratives. We find that attribution methods robustly predict brain activity across the language network, revealing a hierarchical pattern: explanations from early layers align with the brain’s initial language processing stages, while later layers correspond to more advanced stages. Additionally, layers with greater influence on next-word prediction—reflected in higher attribution scores—demonstrate stronger brain alignment. These results underscore XAI’s potential for exploring the neural basis of language and suggest brain alignment for assessing the biological plausibility of explanation methods. © 2025, CC BY-SA.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

Foreword to the 6th International Workshop on Engineering and Cybersecurity of Critical Systems (EnCyCriS)

Foreword to the 6th International Workshop on Engineering an...

引用

Engineering and Cybersecurity of Critical Systems (EnCyCriS), IEEE/ACM International Workshop on

作者： Coralie Esnoul Eunkyoung Jee John Eidar Simensen Ita Ryan Mary Sánchez-Gordón Nathan Lau Risk and Safety Department Institute for Energy Technology Halden Norway School of Computing Korea Advanced Institute of Science and Technology Daejeon Republic of Korea University College Cork Cork Ireland Dept. of Computer Science and Communication Østfold University College Halden Norway Dept. of Industrial and Systems Engineering Virginia Tech Virginia USA

ISBN: (数字)9798331538101

ISBN: (纸本)9798331538118

This marks the 6th edition of the International Workshop on Engineering and Cybersecurity of Critical Systems (EnCyCriS) at the ICSE conference. For industrial critical systems, although the previous premises of increasing system interconnectivity, decentralization, and introduction of new, more intelligent technologies still hold true, there is an increased societal awareness with regards to cybersecurity. This has led to clearer regulation, sharper requirements, and higher expectations for industry. At the same time, the availability of readily deployable competence, methods, tools, and solutions is lacking, which should be considered a critical societal risk. In the current international political climate, cyber security, and safety of critical infrastructures across industry, are more important than ever before. The EnCyCriS workshop facilitates discourse and discussion amongst researchers, practitioners, and students who are working on challenges and solutions related to industrial critical infrastructure. It has a particular focus on sharing industry experience and project results pertaining to cyber threats on critical systems, secure systems engineering, and attack detection and response mechanisms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Perspective-taking for promoting prosocial behaviors through robot-robot VR task

Perspective-taking for promoting prosocial behaviors through...

引用

IEEE International Workshop on Robot and Human Communication (ROMAN)

作者： Hang Chenlin Ono Tetsuo Seiji Yamada Department of Informatics The Graduate University for Advanced Studies Sokendai Kanagawa Japan National Institute of Informatics Tokyo Japan Faculty of Information Science and Technology Division of Computer Science and Information Technology Hokkaido University Sapporo Japan

Perspective-taking, which enables individuals to consider the thoughts and objectives of another, is well established to be a successful strategy for encouraging pro-social behavior in human-computer interactions. Nowadays, perspective-taking is no longer limited to text; it is now more frequently used in virtual reality (VR). However, most previous research has focused on simulating human-human interactions in the real world in VR by providing participants with experiences connected to different moral tasks. In this study, we investigated whether participants’ prosocial behaviors toward robots would change if they experienced an altruistic VR task involving robots from the perspective of different robots. Our findings show that participants who had the help-receiver-view exhibited more altruistic behaviors toward a robot than those who had the help-provider-view one in a dictator game. We believe that this work is the first attempt to investigate the relationship between perspective-taking in a VR environment and changes in prosocial behavior in human-robot interaction.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Real Time Emotion Support System in Text Mining [RTESTM]

Real Time Emotion Support System in Text Mining [RTESTM]

引用

2021 International Conference on computer Communication and Informatics, ICCCI 2021

作者： Ramya, R.S. Sandhya, K. Venugopal, K.R. Iyengar, S.S. Patnaik, L.M. Rns Institute of Technology Bengaluru India University Visvesvaraya College of Engineering Bangalore University Department of Computer Science and Engineering Bengaluru India Bangalore University Bangalore India School of Computing and Information Sciences Florida International University United States National Institute of Advanced Studies Bangalore India

ISBN: (纸本)9781728158754

Mining opinions from online reviews is an essential step in obtaining the overall sentiment of a product. Deep learning procedure is applied over various fields. User ratings are huge for recommender structures since they consolidate various kinds of energetic information that may influence the exactness of the suggestion. In this work, a deep learning model is utilized to process the user remarks and to create a potential user rating for user comments is proposed. To start with, the system uses sentiments to create a feature vector as the input nodes. Further, the framework tools reduce the noise in the dataset to recover the classification of information mining. To finish, Deep Belief Network (DBN) and sentiment analysis reaches data learning for the approvals. © 2021 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Exact Recovery of Two-Latent Variable Stochastic Block Model with Side Information 7

Exact Recovery of Two-Latent Variable Stochastic Block Model...

引用

7th International Conference on Contemporary Information Technology and Mathematics, ICCITM 2021

作者： Shahiri, Mohammad Eskandari, Mahdi Institute for Advanced Studies in Basic Sciences Department of Computer Science and Information Technology Zanjan Iran School of Engineering and Digital Arts University of Kent Canterbury England United Kingdom

ISBN: (纸本)9781665420327

The two-latent variable stochastic block model is a new graph synthetic model making a connection between the conventional stochastic block model and real-world networks. In this model, each node contains two latent variables such that at least one of these two latent variables is unknown. Still, this model lonely is not able to model a real-world network. Side information is another component that sometimes exists beside a real-world network. In this paper, we will investigate the asymptotic behavior of the two-latent variable stochastic block model in the presence of side information. Two different types of side information are considered in this paper: noisy labels and partially revealed labels side information. For each case, the necessary and sufficient conditions for the exact recovery of the desired latent variable are obtained via semidefinite programming optimization. It is shown that these conditions are tight and create a phase transition for the exact recovery. © 2021 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Towards Robust Family-Infant Audio Analysis Based on Unsupervised Pretraining of Wav2vec 2.0 on Large-Scale Unlabeled Family Audio

arXiv

引用

arXiv 2023年

作者： Li, Jialu Hasegawa-Johnson, Mark McElwain, Nancy L. Department of Electrical and Computer Engineering University of Illinois United States Beckman Institute for Advanced Science and Technology University of Illinois United States Department of Human Development and Family Studies University of Illinois United States

To perform automatic family audio analysis, past studies have collected recordings using phone, video, or audio-only recording devices like LENA, investigated supervised learning methods, and used or fine-tuned general-purpose embeddings learned from large pretrained models. In this study, we advance the audio component of a new infant wearable multi-modal device called LittleBeats (LB) by learning family audio representation via wav2vec 2.0 (W2V2) pretraining. We show given a limited number of labeled LB home recordings, W2V2 pretrained using 1k-hour of unlabeled home recordings outperforms oracle W2V2 pretrained on 960-hour unlabeled LibriSpeech in terms of parent/infant speaker diarization (SD) and vocalization classifications (VC) at home. Extra relevant external unlabeled and labeled data further benefit W2V2 pretraining and fine-tuning. With SpecAug and environmental speech corruptions, we obtain 12% relative gain on SD and moderate boost on VC. Code and model weights are available. © 2023, CC BY.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Psychometrics for Hypnopaedia-Aware Machinery via Chaotic Projection of Artificial Mental Imagery

arXiv

引用

arXiv 2024年

作者： Chang, Ching-Chun Gao, Kai Xu, Shuying Kordoni, Anastasia Leckie, Christopher Echizen, Isao National Institute of Informatics Chiyoda Tokyo Japan The Department of Information Engineering and Computer Science Feng Chia University Taichung Taiwan The Department of Psychology Lancaster University Lancaster United Kingdom The Department of Computing and Information Systems University of Melbourne MelbourneVIC Australia The Information and Society Research Division National Institute of Informatics The Department of Informatics Graduate University for Advanced Studies The Department of Information and Communication Engineering University of Tokyo Chiyoda Tokyo Japan

Neural backdoors represent insidious cybersecurity loopholes that render learning machinery vulnerable to unauthorised manipulations, potentially enabling the weaponisation of artificial intelligence with catastrophic consequences. A backdoor attack involves the clandestine infiltration of a trigger during the learning process, metaphorically analogous to hypnopaedia, where ideas are implanted into a subject's subconscious mind under the state of hypnosis or unconsciousness. When activated by a sensory stimulus, the trigger evokes conditioned reflex that directs a machine to mount a predetermined response. In this study, we propose a cybernetic framework for constant surveillance of backdoors threats, driven by the dynamic nature of untrustworthy data sources. We develop a self-aware unlearning mechanism to autonomously detach a machine's behaviour from the backdoor trigger. Through reverse engineering and statistical inference, we detect deceptive patterns and estimate the likelihood of backdoor infection. We employ model inversion to elicit artificial mental imagery, using stochastic processes to disrupt optimisation pathways and avoid convergent but potentially flawed patterns. This is followed by hypothesis analysis, which estimates the likelihood of each potentially malicious pattern being the true trigger and infers the probability of infection. The primary objective of this study is to maintain a stable state of equilibrium between knowledge fidelity and backdoor vulnerability. © 2024, CC BY.

关键词： Reverse engineering

来源：评论

学校读者我要写书评

暂无评论

A Parameter-Efficient Quantum Anomaly Detection Method on a Superconducting Quantum Processor

arXiv

引用

arXiv 2024年

作者： Wang, Maida Jiang, Jinyang Coveney, Peter V. Centre for Computational Science Department of Chemistry University College London United Kingdom Department of Computer Science Tufts University Medford United States Guanghua School of Management Peking University Beijing China Advanced Research Computing Centre University College London United Kingdom Institute for Informatics University of Amsterdam Netherlands

Quantum machine learning has gained attention for its potential to address computational challenges. However, whether those algorithms can effectively solve practical problems and outperform their classical counterparts, especially on current quantum hardware, remains a critical question. In this work, we propose a novel quantum machine learning method, called Parameter-Efficient Quantum Anomaly Detection (PEQAD), for practical image anomaly detection, which aims to achieve both parameter efficiency and superior accuracy compared to classical models. Emulation results indicate that PEQAD demonstrates favourable recognition capabilities compared to classical baselines, achieving an average accuracy of over 90% on benchmarks with significantly fewer trainable parameters. Theoretical analysis confirms that PEQAD has a comparable expressivity to classical counterparts while requiring only a fraction of the parameters. Furthermore, we demonstrate the first implementation of a quantum anomaly detection method for general image datasets on a superconducting quantum processor. Specifically, we achieve an accuracy of over 80% with only 16 parameters on the device, providing initial evidence of PEQAD’s practical viability in the noisy intermediate-scale quantum era and highlighting its significant reduction in parameter requirements. Copyright © 2024, The Authors. All rights reserved.

关键词： Qubits

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：