检索结果-内蒙古大学图书馆

37th Conference on Neural Information Processing Systems, NeurIPS 2023

作者： Xu, Mingyu Lian, Zheng Feng, Lei Liu, Bin Tao, Jianhua The State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China School of Computer Science and Engineering Nanyang Technological University Singapore Department of Automation Tsinghua University China Beijing National Research Center for Information Science and Technology Tsinghua University China

ISBN: (纸本)9781713899921

Noisy partial label learning (noisy PLL) is an important branch of weakly supervised learning. Unlike PLL where the ground-truth label must conceal in the candidate label set, noisy PLL relaxes this constraint and allows the ground-truth label may not be in the candidate label set. To address this challenging problem, most of the existing works attempt to detect noisy samples and estimate the ground-truth label for each noisy sample. However, detection errors are unavoidable. These errors can accumulate during training and continuously affect model optimization. To this end, we propose a novel framework for noisy PLL with theoretical interpretations, called "Adjusting Label Importance Mechanism (ALIM)". It aims to reduce the negative impact of detection errors by trading off the initial candidate set and model outputs. ALIM is a plug-in strategy that can be integrated with existing PLL approaches. Experimental results on multiple benchmark datasets demonstrate that our method can achieve state-of-the-art performance on noisy PLL. Our code is available at: https://***/zeroQiaoba/ALIM. © 2023 Neural information processing systems foundation. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

MSAmba: Exploring Multimodal Sentiment Analysis with State Space Models 39

MSAmba: Exploring Multimodal Sentiment Analysis with State S...

引用

39th Annual AAAI Conference on artificial intelligence, AAAI 2025

作者： He, Xilin Liang, Haijian Peng, Boyi Xie, Weicheng Khan, Muhammad Haris Song, Siyang Yu, Zitong Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing China Mohamed bin Zayed University of Artificial Intelligence United Arab Emirates University of Exeter United Kingdom Great Bay University China

ISBN: (纸本)157735897X

Multimodal sentiment analysis, which learns a model to process multiple modalities simultaneously and predict a sentiment value, is an important area of affective computing. Modeling sequential intra-modal information and enhancing cross-modal interactions are crucial to multimodal sentiment analysis. In this paper, we propose MSAmba, a novel hybrid Mamba-based architecture for multimodal sentiment analysis, consisting of two core blocks: Intra-Modal Sequential Mamba (ISM) block and Cross-Modal Hybrid Mamba (CHM) block, to comprehensively address the above-mentioned challenges with hybrid state space models. Firstly, the ISM block models the sequential information within each modality in a bi-directional manner with the assistance of global information. Subsequently, the CHM blocks explicitly model centralized cross-modal interaction with a hybrid combination of Mamba and attention mechanism to facilitate information fusion across modalities. Finally, joint learning of the intra-modal tokens and cross-modal tokens is utilized to predict the sentiment values. This paper serves as one of the pioneering works to unravel the outstanding performances and great research potential of Mamba-based methods in the task of multimodal sentiment analysis. Experiments on CMU-MOSI, CMU-MOSEI and CH-SIMS demonstrate the superior performance of the proposed MSAmba over prior Transformer-based and CNN-based methods. Copyright © 2025, Association for the Advancement of artificial intelligence (***). All rights reserved.

关键词： Modal analysis

来源：评论

学校读者我要写书评

暂无评论

Improving Complex Reasoning over Knowledge Graph with Logic-Aware Curriculum Tuning 39

Improving Complex Reasoning over Knowledge Graph with Logic-...

引用

39th Annual AAAI Conference on artificial intelligence, AAAI 2025

作者： Xia, Tianle Ding, Liang Wan, Guojia Zhan, Yibing Du, Bo Tao, Dacheng School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China The University of Sydney Australia JD Explore Academy College of Computing and Data Science Nanyang Technological University 639798 Singapore

ISBN: (纸本)157735897X

Answering complex queries over incomplete knowledge graphs (KGs) is a challenging task. Most previous works have focused on learning entity/relation embeddings and simulating first-order logic operators with various neural networks. However, they are bottlenecked by the inability to share world knowledge to improve logical reasoning, thus resulting in suboptimal performance. In this paper, we propose a complex reasoning schema over KG upon large language models (LLMs), containing a curriculum-based logical-aware instruction tuning framework, named LACT. Specifically, we augment the arbitrary first-order logical queries via binary tree decomposition, to stimulate the reasoning capability of LLMs. To address the difficulty gap among different types of complex queries, we design a simple and flexible logic-aware curriculum learning framework. Experiments across widely used datasets demonstrate that LACT has substantial improvements (brings an average +5.5% MRR score) over advanced methods, achieving the new state-of-the-art. Copyright © 2025, Association for the Advancement of artificial intelligence (***). All rights reserved.

关键词： Curricula

来源：评论

学校读者我要写书评

暂无评论

CA-MLIF: Cross-Attention and Multimodal Low-Rank Interaction Fusion Framework for Tumor Prognostic Prediction 39

CA-MLIF: Cross-Attention and Multimodal Low-Rank Interaction...

引用

39th Annual AAAI Conference on artificial intelligence, AAAI 2025

作者： An, Yajun Chen, Jiale Lin, Huan Liu, Zhenbing Feng, Siyang Zhang, Hualong Lan, Rushi Liu, Zaiyi Pan, Xipeng School of Computer Science and Information Security Guilin University of Electronic Technology Guilin541004 China School of Artificial Intelligence and Big Data Chongqing College of Finance and Economics Chongqing402160 China Southern Medical University Guangzhou510080 China Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin541004 China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangzhou510080 China

ISBN: (纸本)157735897X

Cancer is a leading cause of death worldwide due to its aggressive nature and complex variability. Accurate prognosis is therefore challenging but essential for guiding personalized treatment and follow-up. Previous research often relied on single data sources, missing the opportunity to combine various types of patient information for more comprehensive survival predictions. To address these challenges, we propose a two-stage fusion method named Cross-Attention and Multimodal Low-Rank Interaction Fusion Framework (CA-MLIF). In the first stage, we propose a CA mechanism for real-time feature updates and cross-modal mutual learning to capture rich semantic information. In the second stage, we design a novel multimodal low-rank interaction fusion method for survival prediction. Specifically, we present modal attention mechanism (MAM) for feature filtration, low-rank multimodal fusion (LMF) for model complexity reduction, and optimal weight concatenation (OWC) for maximizing feature integration. Extensive experiments on two public datasets TCGA-GBMLGG and TCGA-KIRC, as well as a multi-center in-house lung adenocarcinoma (LUAD) dataset validate the effectiveness of CA-MLIF, which demonstrate that our method outperforms existing approaches in survival prediction under both pathology-gene fusion and CT-pathology fusion scenarios. Copyright © 2025, Association for the Advancement of artificial intelligence (***). All rights reserved.

关键词： Learning to rank

来源：评论

学校读者我要写书评

暂无评论

A Preliminary Study on the Explicitness of Bug Associations 34

A Preliminary Study on the Explicitness of Bug Associations

引用

34th International Conference on Software Engineering and Knowledge Engineering, SEKE 2022

作者： Li, Zengyang Xu, Jieling Cai, Guangzong Liang, Peng Mo, Ran School of Computer Science Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning Central China Normal University Wuhan China School of Computer Science Wuhan University Wuhan China

ISBN: (纸本)1891706543

Bugs are usually in associations with other bugs in a software system, e.g., a bug may result from another bug. However, such bug associations are implicit and usually cannot be traced without a significant amount of effort. Intuitively, if a bug association is easier to trace, the involved bugs can be fixed in a cleaner way. However, there is little evidence on the explicitness of bug associations. In this paper, we aim to evaluate the explicitness of bug associations, so as to get a basic understanding on such associations. To this end, we defined a metric to quantify the explicitness of a bug association, and conducted an empirical study on 11 non-trivial Apache open source software systems. The main findings are summarized as follows: (1) From the perspective of code change history, around 29% of bug pairs are not explicitly associated, and about 71% are explicitly associated to some extent;(2) Bugs in the association of Container have relatively strong association explicitness, while bugs in the association of Blocked or Blocker, Cloners, and Dependent have relatively weak association explicitness. These findings provide insights on software analyzability to practitioners and researchers. © 2022 Knowledge Systems Institute Graduate School. All rights reserved.

关键词： Open source software

来源：评论

学校读者我要写书评

暂无评论

Probability-Based Channel Pruning for Depthwise Separable Convolutional Networks

引用

Journal of computer science & Technology 2022年第3期37卷 584-600页

作者： Han-Li Zhao Kai-Jie Shi Xiao-Gang Jin Ming-Liang Xu Hui Huang Wang-Long Lu Ying Liu College of Computer Science and Artificial Intelligence Wenzhou UniversityWenzhou 325035China State Key Laboratory of CAD&CG Zhejiang UniversityHangzhou 310058China School of Information Engineering Zhengzhou UniversityZhengzhou 450000China Department of Computer Science Memorial University of NewfoundlandSt.John's A1B 3X5Canada

Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network ***,existing channel pruning methods mainly focus on the pruning of standard convolutional networks,and they rely intensively on time-consuming fine-tuning to achieve the performance *** this end,we present a novel efficient probability-based channel pruning method for depthwise separable convolutional *** method leverages a new simple yet effective probability-based channel pruning criterion by taking the scaling and shifting factors of batch normalization layers into consideration.A novel shifting factor fusion technique is further developed to improve the performance of the pruned networks without requiring extra time-consuming *** apply the proposed method to five representative deep learning networks,namely MobileNetV1,MobileNetV2,ShuffleNetV1,ShuffleNetV2,and GhostNet,to demonstrate the efficiency of our pruning *** experimental results and comparisons on publicly available CIFAR10,CIFAR100,and ImageNet datasets validate the feasibility of the proposed method.

关键词： network compression channel pruning depthwise separable convolution batch normalization

来源：评论

学校读者我要写书评

暂无评论

Activation Template Matching Loss for Explainable Face Recognition 17

Activation Template Matching Loss for Explainable Face Recog...

引用

17th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2023

作者： Lin, Huawei Liu, Haozhe Li, Qiufu Shen, Linlin Computer Vision Institute School of Computer Science and Softwre Enginnering Shenzhen University Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China

ISBN: (纸本)9798350345445

Can we construct an explainable face recognition network able to learn a facial part-based feature like eyes, nose, mouth and so forth, without any manual annotation or additionalsion datasets? In this paper, we propose a generic Explainable Channel Loss (ECLoss) to construct an explainable face recognition network. The explainable network trained with ECLoss can easily learn the facial part-based representation on the target convolutional layer, where an individual channel can detect a certain face part. Our experiments on dozens of datasets show that ECLoss achieves superior explainability metrics, and at the same time improves the performance of face verification without face alignment. In addition, our visualization results also illustrate the effectiveness of the proposed ECLoss. © 2023 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

artificial Impostors: An Efficient and Scalable Scheme for Location Privacy Preservation

引用

IEEE Transactions on Services Computing 2025年第3期18卷 1262-1277页

作者： Hao Tang Kunfeng Chen Zhiyang Xie Cheng Wang College of Computer Science and Technology Tongji University Shanghai China Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai China Shanghai Artificial Intelligence Laboratory Shanghai China

The progress of location-based services has led to severe concerns about location privacy leakage. However, existing methods are still incompetent for efficient and scalable location privacy preservation (LPP). They are often vulnerable to inference attacks with side information, or hard to be implemented due to the high computational complexity. In this article, we pursue the high protection quality with low computational complexity. We propose a scalable LPP method based on the paradigm of counterfeiting locations. To make fake locations extremely plausible, we forge them by synthesizing artificial impostors. The so-called artificial impostors refer to the synthesized traces that have similar semantic features to the actual traces, e.g., similar transition patterns, and do not contain any protected location, i.e., the exact location that needs to be protected. We devise two dedicated techniques: the station-based synthesis method and the population-level semantic model. We conduct the experiments on real datasets of two cities (Shanghai of China and Asturias of Spain) to validate the quality of privacy preservation, utility loss, and scalability of the proposed method. Based on these two datasets, the experimental results show that our method achieves the privacy preservation quality of 97.68% and 96.24%, respectively, and the time spent on building generators is only 144.96 seconds and 136.08 seconds, respectively. The experimental results also show that our method achieves a good trade-off between privacy and utility. Our study would give the research community new insights into improving the practicality of LPP paradigm via counterfeiting locations.

关键词： Privacy Semantics Feature extraction Urban areas Generators Protection Counterfeiting Computational efficiency artificial intelligence Transforms

来源：评论

学校读者我要写书评

暂无评论

MDDMamba: A Model for Multi-Modal Depression Detection with a Memory-Saving Cross-Modal Attention Mechanism Based on Mamba

MDDMamba: A Model for Multi-Modal Depression Detection with ...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Liu, Mingzhou Wang, Zheyu Liu, Feng Nanjing University of Posts and Telecommunications School of Education Science and Technology Nanjing210023 China Nanjing University of Science and Technology ZiJin College School of Computer and Artificial Intelligence Nanjing210023 China Jiangsu Key Laboratory of Intelligent Information Processing and Communication Technology Nanjing210003 China

ISBN: (纸本)9798350386226

Depression is a severe mental health disorder that constitutes a significant global health challenge, necessitating efficient and accurate automated diagnostic tools. Traditional approaches to automated depression assessment using multi-modal data often face three significant difficulties: small sample sizes in available datasets, high computational cost, and difficulties in effectively integrating different modalities. In this paper, we first mitigate the problem of dataset quality by utilizing an estimated imputation strategy for missing values based on a known data augmentation method. Additionally, we introduce a multi-modal depression detection model named MDDMamba, designed to address the latter two challenges by integrating features from pre-trained emotion models with the memory-efficient selective state space model named Mamba. The MDDMamba model combines features from pre-trained emotion models for both text and audio, enhancing the extraction of emotional information. The core of the MDDMamba model is the CrossMamba module, which replaces the traditional Transformer's quadratic computational complexity with a linear complexity alternative, facilitating efficient multi-modal fusion. Experimental results on CMDC and EATD-Corpus datasets show that the MDDMamba model achieves better classification and regression performance and stronger generalization capabilities in depression detection. Notably, the MDDMamba model achieves an F1 score of 0.94 and a Pearson correlation coefficient of 0.86 on the CMDC dataset while maintaining lower memory cost, outperforming most advanced multi-modal depression detection methods across various experimental scenarios. © 2024 IEEE.

关键词： Data fusion

来源：评论

学校读者我要写书评

暂无评论

MASKED DISTILLATION ADVANCES SELF-SUPERVISED TRANSFORMER ARCHITECTURE SEARCH 12

MASKED DISTILLATION ADVANCES SELF-SUPERVISED TRANSFORMER ARC...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Yan, Caixia Chang, Xiaojun Li, Zhihui Yao, Lina Luo, Minnan Zheng, Qinghua School of Computer Science and Technology MOEKLINNS Laboratory Xi'an Jiaotong University China Mohamed bin Zayed University of Artificial Intelligence United Arab Emirates School of Information Science and Technology University of Science and Technology of China China CSIRO's Data61 University of New South Wales Australia

Transformer architecture search (TAS) has achieved remarkable progress in automating the neural architecture design process of vision transformers. Recent TAS advancements have discovered outstanding transformer architectures while saving tremendous labor from human experts. However, it is still cumbersome to deploy these methods in real-world applications due to the expensive costs of data labeling under the supervised learning paradigm. To this end, this paper proposes a masked image modelling (MIM) based self-supervised neural architecture search method specifically designed for vision transformers, termed as MaskTAS, which completely avoids the expensive costs of data labeling inherited from supervised learning. Based on the one-shot NAS framework, MaskTAS requires to train various weight-sharing subnets, which can easily diverged without strong supervision in MIM-based self-supervised learning. For this issue, we design the search space of MaskTAS as a siamesed teacher-student architecture to distill knowledge from pre-trained networks, allowing for efficient training of the transformer supernet. To achieve self-supervised transformer architecture search, we further design a novel unsupervised evaluation metric for the evolutionary search algorithm, where each candidate of the student branch is rated by measuring its consistency with the larger teacher network. Extensive experiments demonstrate that the searched architectures can achieve state-of-the-art accuracy on CIFAR-10, CIFAR-100, and ImageNet datasets even without using manual labels. Moreover, the proposed MaskTAS can generalize well to various data domains and tasks by searching specialized transformer architectures in self-supervised manner. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Supervised learning

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：