检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 1411-1424页

作者： Zhang, Yuhong Shi, Congmei Li, Xinzheng Zhang, Zan Hu, Xuegang School of Computer Science and Information Engineering Hefei University of Technology Hefei230601 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230088 China Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Ministry of Education Hefei230009 China

Cross-network node classification aims to train a classifier for an unlabeled target network using a source network with rich labels. In applications, the degree of nodes mostly conforms to the long-tail distribution, i.e., most nodes in the network are tail nodes with sparse neighborhoods. The established methods focus on either the discrepancy cross network or the long tail in a single network. As for the cross-network node classification under long tail, the coexistence of sparsity of tail nodes and the discrepancy cross-network challenges existing methods for long tail or methods for the cross-network node classification. To this end, a multicomponent similarity graphs for cross-network node classification (MS-CNC) is proposed in this article. Specifically, in order to address the sparsity of the tail nodes, multiple component similarity graphs, including attribute and structure similarity graphs, are constructed for each network to enrich the neighborhoods of the tail nodes and alleviate the long-tail phenomenon. Then, multiple representations are learned from the multiple similarity graphs separately. Based on the multicomponent representations, a two-level adversarial model is designed to address the distribution difference across networks. One level is used to learn the invariant representations cross network in view of structure and attribute components separately, and the other level is used to learn the invariant representations in view of the fused structure and attribute graphs. Extensive experimental results show that the MS-CNC outperforms the state-of-the-art methods. Impact Statement-Node classification is an important task in graph mining. With the unavailability of labels, some researchers propose cross-network node classification, using one labeled network to assist the node classification of another unlabeled network. However, the long-tail of nodes leads to unsatisfactory performance and challenges the recent cross-network node classification m

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Hypernetwork-Assisted Parameter-Efficient Fine-Tuning with Meta-Knowledge Distillation for Domain Knowledge Disentanglement

Hypernetwork-Assisted Parameter-Efficient Fine-Tuning with M...

引用

2024 Findings of the Association for Computational Linguistics: NAACL 2024

作者： Li, Changqun Wang, Linlin Lin, Xin Huang, Shizhou He, Liang School of Computer Science and Technology East China Normal University China Shanghai Key Laboraiory of Multdimensional Infomation Procesing China Shanghai Artificial Intelligence Laboratory China

ISBN: (纸本)9798891761193

Domain adaptation from labeled source domains to the target domain is important in practical summarization scenarios. However, the key challenge is domain knowledge disentanglement. In this work, we explore how to disentangle domain-invariant knowledge from source domains while learning specific knowledge of the target domain. Specifically, we propose a hypernetwork-assisted encoder-decoder architecture with parameter-efficient fine-tuning. It leverages a hypernetwork instruction learning module to generate domain-specific parameters from the encoded inputs accompanied by task-related instruction. Further, to better disentangle and transfer knowledge from source domains to the target domain, we introduce a meta-knowledge distillation strategy to build a meta-teacher model that captures domain-invariant knowledge across multiple domains and use it to transfer knowledge to students. Experiments on three dialogue summarization datasets show the effectiveness of the proposed model. Human evaluations also show the superiority of our model with regard to the summary generation quality. © 2024 Association for Computational Linguistics.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Dual-Modeling Decouple Distillation for Unsupervised Anomaly Detection 24

Dual-Modeling Decouple Distillation for Unsupervised Anomaly...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Liu, Xinyue Wang, Jianyuan Leng, Biao Zhang, Shuo School of Computer Science and Engineering Beihang University Beijing China Key Laboratory of Intelligent Bionic Unmanned Systems Ministry of Education School of Intelligence Science and Technology University of Science and Technology Beijing Beijing China Beijing Key Lab of Traffic Data Analysis and Mining School of Computer & Technology Beijing Jiaotong University Beijing China

ISBN: (纸本)9798400706868

Knowledge distillation based on student-teacher network is one of the mainstream solution paradigms for the challenging unsupervised Anomaly Detection task, utilizing the difference in representation capabilities of the teacher and student networks to implement anomaly localization. However, over-generalization of the student network to the teacher network may lead to negligible differences in representation capabilities of anomaly, thus affecting the detection effectiveness. Existing methods address the possible over-generalization by using differentiated students and teachers from the structural perspective or explicitly expanding distilled information from the content perspective, which inevitably results in an increased likelihood of underfitting of the student network and poor anomaly detection capabilities in anomaly center or edge. In this paper, we propose Dual-Modeling Decouple Distillation (DMDD) for the unsupervised Anomaly Detection. In DMDD, a Decouple Student-Teacher Network is proposed to decouple the initial student features into normality and abnormality features. We further introduce Dual-Modeling Distillation based on normal-anomalous image pairs, fitting normality features of anomalous image and the teacher features of the corresponding normal image, widening the distance between abnormality features and the teacher features in anomalous regions. Synthesizing these two distillation ideas, we achieve anomaly detection which focuses on both edge and center of anomaly. Finally, a Multi-perception Segmentation Network is proposed to achieve focused anomaly map fusion based on multiple attention. Experimental results on MVTec AD show that DMDD surpasses SOTA localization performance of previous knowledge distillation-based methods, reaching 98.85% on pixel-level AUC and 96.13% on PRO. © 2024 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

ControlSynth Neural ODEs: Modeling Dynamical Systems with Guaranteed Convergence 38

ControlSynth Neural ODEs: Modeling Dynamical Systems with Gu...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Mei, Wenjie Zheng, Dongzhe Li, Shihua The School of Automation and the Key Laboratory of MCCSE of the Ministry of Education Southeast University Nanjing China The Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China

Neural ODEs (NODEs) are continuous-time neural networks (NNs) that can process data without the limitation of time intervals. They have advantages in learning and understanding the evolution of complex real dynamics. Many previous works have focused on NODEs in concise forms, while numerous physical systems taking straightforward forms, in fact, belong to their more complex quasi-classes, thus appealing to a class of general NODEs with high scalability and flexibility to model those systems. This, however, may result in intricate nonlinear properties. In this paper, we introduce ControlSynth Neural ODEs (CSODEs). We show that despite their highly nonlinear nature, convergence can be guaranteed via tractable linear inequalities. In the composition of CSODEs, we introduce an extra control term for learning the potential simultaneous capture of dynamics at different scales, which could be particularly useful for partial differential equation-formulated systems. Finally, we compare several representative NNs with CSODEs on important physical dynamics under the inductive biases of CSODEs, and illustrate that CSODEs have better learning and predictive abilities in these settings. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mixture of Experts for Audio-Visual Learning 38

Mixture of Experts for Audio-Visual Learning

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Cheng, Ying Li, Yang He, Junjie Feng, Rui School of Computer Science Fudan University China Shanghai Key Laboratory of Intelligent Information Processing China Shanghai Collaborative Innovation Center of Intelligent Visual Computing China

With the rapid development of multimedia technology, audio-visual learning has emerged as a promising research topic within the field of multimodal analysis. In this paper, we explore parameter-efficient transfer learning for audio-visual learning and propose the Audio-Visual Mixture of Experts (AVMoE) to inject adapters into pre-trained models flexibly. Specifically, we introduce unimodal and cross-modal adapters as multiple experts to specialize in intra-modal and inter-modal information, respectively, and employ a lightweight router to dynamically allocate the weights of each expert according to the specific demands of each task. Extensive experiments demonstrate that our proposed approach AVMoE achieves superior performance across multiple audio-visual tasks, including AVE, AVVP, AVS, and AVQA. Furthermore, visual-only experimental results also indicate that our approach can tackle challenging scenes where modality information is missing. The source code is available at https://***/yingchengy/AVMOE. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

P-ICL: Point In-Context Learning for Named Entity Recognition with Large Language Models

arXiv

引用

arXiv 2024年

作者： Jiang, Guochao Ding, Zepeng Shi, Yuchen Yang, Deqing School of Data Science Fudan University Shanghai China Shanghai Key Laboratory of Data Science Shanghai China

In recent years, the rise of large language models (LLMs) has made it possible to directly achieve named entity recognition (NER) without any demonstration samples or only using a few samples through in-context learning (ICL). However, standard ICL only helps LLMs understand task instructions, format and input-label mapping, but neglects the particularity of the NER task itself. In this paper, we propose a new prompting framework P-ICL to better achieve NER with LLMs, in which some point entities are leveraged as the auxiliary information to recognize each entity type. With such significant information, the LLM can achieve entity classification more precisely. To obtain optimal point entities for prompting LLMs, we also proposed a point entity selection method based on K-Means clustering. Our extensive experiments on some representative NER benchmarks verify the effectiveness of our proposed strategies in P-ICL and point entity selection.1 © 2024, CC BY-SA.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Bipartite synchronization of stochastic coupled systems with hybrid time-varying delays via asynchronous impulsive control

引用

IEEE Transactions on Automation science and Engineering 2025年 22卷 15778-15791页

作者： Liu, Yan Sun, Yiliang Zhou, Hui Tiangong University Tianjin Key Laboratory of Autonomous Intelligence Technology and Systems School of Computer Science and Technology Tianjin 300387 China Harbin Institute of Technology (Weihai) Department of Mathematics Weihai 264209 China Fuzhou University College of Computer and Data Science Fuzhou 350108 China

This paper investigates the bipartite synchronization of stochastic coupled systems with hybrid time-varying delays and Markov jump via asynchronous impulsive control. Unlike existing studies, the asynchronous impulses considered in this paper are decentralized, which allows for the investigation of a wider range of dynamic properties. Additionally, to address asynchronous impulsive control, asynchronous average dwell-time condition is introduced for the first time, and new auxiliary functions are defined. Several sufficient criteria for attaining bipartite synchronization are derived through the application of the Lyapunov method and graph theory. In addition, the theoretical findings are applied to single-link robot arms with hybrid time-varying delays. To demonstrate the practicality of the findings, this paper includes two numerical examples. © 2004-2012 IEEE.

关键词： asynchronous impulsive control bipartite synchronization hybrid time-varying delays single-link robot arms

来源：评论

学校读者我要写书评

暂无评论

Reason from Fallacy: Enhancing Large Language Models’ Logical Reasoning through Logical Fallacy Understanding

arXiv

引用

arXiv 2024年

作者： Li, Yanda Wang, Dixuan Liang, Jiaqing Jiang, Guochao He, Qianyu Xiao, Yanghua Yang, Deqing School of Data Science Fudan University Shanghai China Shanghai Key Laboratory of Data Science Shanghai China

Large Language Models (LLMs) have demonstrated good performance in many reasoning tasks, but they still struggle with some complicated reasoning tasks including logical reasoning. One non-negligible reason for LLMs’ suboptimal performance on logical reasoning is their overlooking of understanding logical fallacies correctly. To evaluate LLMs’ capability of logical fallacy understanding (LFU), we propose five concrete tasks from three cognitive dimensions of WHAT, WHY, and HOW in this paper. Towards these LFU tasks, we have successfully constructed a new dataset LFUD based on GPT-4 accompanied by a little human effort. Our extensive experiments justify that our LFUD can be used not only to evaluate LLMs’ LFU capability, but also to fine-tune LLMs to obtain significantly enhanced performance on logical reasoning. © 2024, CC BY.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Representation learning via an integrated autoencoder for unsupervised domain adaptation

引用

Frontiers of computer science 2023年第5期17卷 75-87页

作者： Yi ZHU Xindong WU Jipeng QIANG Yunhao YUAN Yun LI School of Information Engineering Yangzhou UniversityYangzhou 225127China Key Laboratory of Knowledge Engineering with Big Data(Ministry of Education of China) Hefei University of TechnologyHefei 230009China School of Computer Science and Information Engineering Hefei University of TechnologyHefei 230601China

The purpose of unsupervised domain adaptation is to use the knowledge of the source domain whose data distribution is different from that of the target domain for promoting the learning task in the target *** key bottleneck in unsupervised domain adaptation is how to obtain higher-level and more abstract feature representations between source and target domains which can bridge the chasm of domain ***,deep learning methods based on autoencoder have achieved sound performance in representation learning,and many dual or serial autoencoderbased methods take different characteristics of data into consideration for improving the effectiveness of unsupervised domain ***,most existing methods of autoencoders just serially connect the features generated by different autoencoders,which pose challenges for the discriminative representation learning and fail to find the real cross-domain *** address this problem,we propose a novel representation learning method based on an integrated autoencoders for unsupervised domain adaptation,called *** capture the inter-and inner-domain features of the raw data,two different autoencoders,which are the marginalized autoencoder with maximum mean discrepancy(mAE)and convolutional autoencoder(CAE)respectively,are proposed to learn different feature *** higher-level features are obtained by these two different autoencoders,a sparse autoencoder is introduced to compact these inter-and inner-domain *** addition,a whitening layer is embedded for features processed before the mAE to reduce redundant features inside a local *** results demonstrate the effectiveness of our proposed method compared with several state-of-the-art baseline methods.

关键词： unsupervised domain adaptation representation learning marginalized autoencoder convolutional autoen-coder sparse autoencoder

来源：评论

学校读者我要写书评

暂无评论

Quantifying and Mitigating Unimodal Biases in Multimodal Large Language Models: A Causal Perspective

Quantifying and Mitigating Unimodal Biases in Multimodal Lar...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Chen, Meiqi Cao, Yixin Zhang, Yan Lu, Chaochao State Key Laboratory of General Artificial Intelligence Peking University Beijing China School of Intelligence Science and Technology Peking University China School of Computer Science Fudan University China Shanghai Artificial Intelligence Laboratory China

ISBN: (纸本)9798891761681

Recent advancements in Large Language Models (LLMs) have facilitated the development of Multimodal LLMs (MLLMs). Despite their impressive capabilities, MLLMs often suffer from over-reliance on unimodal biases (e.g., language bias and vision bias), leading to incorrect answers in complex multimodal tasks. To investigate this issue, we propose a causal framework to interpret the biases in Visual Question Answering (VQA) problems. Within this framework, we conduct an in-depth causal analysis to assess the causal effect of these biases on MLLM predictions. Based on the analysis, we introduce 1) a novel MORE dataset with 12,000 challenging VQA instances requiring multi-hop reasoning and overcoming unimodal biases. 2) a causality-enhanced agent framework CAVE that guides models to comprehensively integrate information from different modalities and mitigate biases. Our experiments show that MLLMs perform poorly on MORE, indicating strong unimodal biases and limited semantic understanding. However, when integrated with our CAVE, promising improvements in reasoning and bias mitigation can be seen. These findings provide important insights for the development of more robust MLLMs and contribute to the broader goal of advancing multimodal AI systems capable of deeper understanding and reasoning. Our project page is at https://***/OpenCausaLab/MORE. © 2024 Association for Computational Linguistics.

关键词： Semantics

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：