检索结果-内蒙古大学图书馆

Proceedings of the 37th International Conference on Neural Information Processing systems

作者： Haochen Wang Junsong Fan Yuxi Wang Kaiyou Song Tong Wang Zhaoxiang Zhang Center for Research on Intelligent Perception and Computing (CRIPAC) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences (CASIA) and University of Chinese Academy of Sciences (UCAS) Center for Research on Intelligent Perception and Computing (CRIPAC) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences (CASIA) and Centre for Artificial Intelligence and Robotics HKISI_CAS Megvii Technology Center for Research on Intelligent Perception and Computing (CRIPAC) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences (CASIA) and University of Chinese Academy of Sciences (UCAS) and Centre for Artificial Intelligence and Robotics HKISI_CAS

As it is empirically observed that Vision Transformers (ViTs) are quite insensitive to the order of input tokens, the need for an appropriate self-supervised pretext task that enhances the location awareness of ViTs is becoming evident. To address this, we present DropPos, a novel pretext task designed to reconstruct Dropped Positions. The formulation of DropPos is simple: we first drop a large random subset of positional embeddings and then the model classifies the actual position for each non-overlapping patch among all possible positions solely based on their visual appearance. To avoid trivial solutions, we increase the difficulty of this task by keeping only a subset of patches visible. Additionally, considering there may be different patches with similar visual appearances, we propose position smoothing and attentive reconstruction strategies to relax this classification problem, since it is not necessary to reconstruct their exact positions in these cases. Empirical evaluations of DropPos show strong capabilities. DropPos outperforms supervised pre-training and achieves competitive results compared with state-of-the-art self-supervised alternatives on a wide range of downstream benchmarks. This suggests that explicitly encouraging spatial reasoning abilities, as DropPos does, indeed contributes to the improved location awareness of ViTs. The code is publicly available at https://***/Haochen-Wang409/DropPos.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards interpretable deep neural networks by leveraging adversarial examples

arXiv

引用

arXiv 2019年

作者： Dong, Yinpeng Bao, Fan Su, Hang Zhu, Jun Tsinghua National Lab for Information Science and Technology State Key Lab of Intelligent Technology and Systems Center for Bio-Inspired Computing Research Department of Computer Science and Technology Tsinghua University Beijing100084 China

Sometimes it is not enough for a DNN to produce an outcome. For example, in applications such as healthcare, users need to understand the rationale of the decisions. Therefore, it is imperative to develop algorithms to learn models with good interpretability (Doshi-Velez 2017). An important factor that leads to the lack of interpretability of DNNs is the ambiguity of neurons, where a neuron may fire for various unrelated concepts. This work aims to increase the interpretability of DNNs on the whole image space by reducing the ambiguity of neurons. In this paper, we make the following contributions: • We propose a metric to evaluate the consistency level of neurons in a network quantitatively. • We find that the learned features of neurons are ambiguous by leveraging adversarial examples. • We propose to improve the consistency of neurons on adversarial example subset by an adversarial training algorithm with a consistent loss. Copyright © 2019, The Authors. All rights reserved.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

Max-Margin Majority Voting for Learning from Crowds 15

Max-Margin Majority Voting for Learning from Crowds

引用

Annual Conference on Neural Information Processing systems

作者： Tian Tian Jun Zhu Department of Computer Science & Technology Center for Bio-Inspired Computing Research Tsinghua National Lab for Information Science & Technology State Key Lab of Intelligent Technology & Systems Tsinghua University Beijing 100084 China

ISBN: (纸本)9781510825024

Learning-from-crowds aims to design proper aggregation strategies to infer the unknown true labels from the noisy labels provided by ordinary web workers. This paper presents max-margin majority voting (M~3V) to improve the discriminative ability of majority voting and further presents a Bayesian generalization to incorporate the flexibility of generative methods on modeling noisy observations with worker confusion matrices. We formulate the joint learning as a regularized Bayesian inference problem, where the posterior regularization is derived by maximizing the margin between the aggregated score of a potential true label and that of any alternative label. Our Bayesian model naturally covers the Dawid-Skene estimator and M~3V. Empirical results demonstrate that our methods are competitive, often achieving better results than state-of-the-art estimators.

关键词： Vote Bayesian inference Crowding Learning Noise Flexibility

来源：评论

学校读者我要写书评

暂无评论

Modeling and analysis of information dissemination mechanism of social media

Modeling and analysis of information dissemination mechanism...

引用

2012 IEEE International Conference on Service Operations and Logistics, and Informatics, SOLI 2012

作者： Hou, Jiachen Xiong, Gang Fan, Dong Nyberg, Timo R. Dongguan Research Institute of CASIA Cloud Computing Center Chinese Academy of Sciences Songshan Lake Dongguan 523808 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China Jiang Su China Sciences Intelligent Engineering Co. Ltd. Enterprise Academician Workstation 215000 China BIT Research Centre Aalto University Aalto Finland

ISBN: (纸本)9781467324007

With the fast development of Internet and data volume increase in Internet, Internet is playing more important role in information spreading in recent years. It is reported that social network is a typical complex network expressed with scale free degree. Emergency cases are happening more and more often in China and all over the world, the governments has realized that it is necessary to monitor and manage the information spreading on social network. In this paper, a new experiment platform frame, which is based on SPA (Spreading Activation) model and social centrality, is proposed to validate the information spreading speed in social network, to help the administration department to manage social network more efficiently. © 2012 IEEE.

关键词： Information dissemination

来源：评论

学校读者我要写书评

暂无评论

Multi-scale cyclical similarity prototype refinement for few-shot breast ultrasound image segmentation

Multi-scale cyclical similarity prototype refinement for few...

引用

International Conference on Signal Processing Proceedings (ICSP)

作者： Yingfeng Ou Xing Yang Jian Zhang Caiqing Jian Lihui Wang Engineering Research Center of Text Computing Ministry of Education Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang China

ISBN: (数字)9798350387384

ISBN: (纸本)9798350387391

Few-shot learning based methods can address the reliance on large-scale labeled samples in current breast tumor segmentation. However, previous methods typically rely on a few support samples to extract abstract, coarse-grained semantic prototypes, without considering the interaction between support and query data. In this study, we aim to provide each query image with relevant and information-rich support information. By integrating the difficult point selection module and the multi-scale cyclical prototype refinement module, we progressively refine coarse-grained support prototypes into more informative, fine-grained class prototypes, thereby better matching the highly variable breast ultrasound query data. Experiments on two datasets demonstrate that our proposed multi-scale cyclical similarity prototype refinement network (CSPRNet) achieves remarkable improvements compared to previous state-of-the-art methods. Specifically, using ResNet50 as the backbone, we achieve 57.93%/57.92% and 57.58%/59.24% mIoU for 1-shot and 5-shot segmentation, respectively, on public and private datasets, outperforming the suboptimal method by 8.18%/2.55% and 2.77%/4.48%.

关键词： Image segmentation Ultrasonic imaging Breast tumors Semantics Prototypes Signal processing Data mining Few shot learning Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Text-Guided Molecule Generation with Diffusion Language Model

arXiv

引用

arXiv 2024年

作者： Gong, Haisong Liu, Qiang Wu, Shu Wang, Liang Center for Research on Intelligent Perception and Computing 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

Text-guided molecule generation is a task where molecules are generated to match specific textual descriptions. Recently, most existing SMILES-based molecule generation methods rely on an autoregressive architecture. In this work, we propose the Text-Guided Molecule Generation with Diffusion Language Model (TGM-DLM), a novel approach that leverages diffusion models to address the limitations of autoregressive methods. TGM-DLM updates token embeddings within the SMILES string collectively and iteratively, using a two-phase diffusion generation process. The first phase optimizes embeddings from random noise, guided by the text description, while the second phase corrects invalid SMILES strings to form valid molecular representations. We demonstrate that TGM-DLM outperforms MolT5-Base, an autoregressive model, without the need for additional data resources. Our findings underscore the remarkable effectiveness of TGM-DLM in generating coherent and precise molecules with specific properties, opening new avenues in drug discovery and related scientific domains. Code will be released at: https://***/Deno-V/tgm-dlm. Copyright © 2024, The Authors. All rights reserved.

关键词： Molecules

来源：评论

学校读者我要写书评

暂无评论

Optimal simulation of Deutsch gates and the Fredkin gate

引用

Physical Review A 2015年第3期91卷 032302-032302页

作者： Nengkun Yu Mingsheng Ying Institute for Quantum Computing University of Waterloo Waterloo Ontario Canada Department of Mathematics & Statistics University of Guelph Guelph Ontario Canada Center for Quantum Computation and Intelligent Systems (QCIS) Faculty of Engineering and Information Technology University of Technology Sydney NSW 2007 Australia State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing 100084 China

In this paper, we study the optimal simulation of the three-qubit unitary using two-qubit gates. First, we completely characterize the two-qubit gate cost of simulating the Deutsch gate (controlled-controlled gate) by generalizing our result on the two-qubit cost of the Toffoli gate. The function of any Deutsch gate is simply a three-qubit controlled-unitary gate and can be intuitively explained as follows: The gate outputs the states of the two control qubits directly, and applies the given one-qubit unitary u on the target qubit only if both the states of the control qubits are |1〉. Previously, it was only known that five two-qubit gates are sufficient for implementing such a gate [Sleator and Weinfurter, Phys. Rev. Lett. 74, 4087 (1995)]. We show that if the determinant of u is 1, four two-qubit gates are optimal. Otherwise, five two-qubit gates are required. For the Fredkin gate (the controlled-swap gate), we prove that five two-qubit gates are necessary and sufficient, which settles the open problem introduced in Smolin and DiVincenzo [Phys. Rev. A 53, 2855 (1996)].

关键词： QUANTUM gates PARTICLE interactions QUBITS QUANTUM information science QUANTUM fluctuations HAMILTONIAN systems QUANTUM states

来源：评论

学校读者我要写书评

暂无评论

Interpretable Multimodal Out-of-Context Detection with Soft Logic Regularization

Interpretable Multimodal Out-of-Context Detection with Soft ...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Huanhuan Ma Jinghao Zhang Qiang Liu Shu Wu Liang Wang School of Artificial Intelligence University of Chinese Academy of Sciences Center for Research on Intelligent Perception and Computing (CRIPAC) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences

The rapid spread of information through mobile devices and media has led to the widespread of false or deceptive news, causing significant concerns in society. Among different types of misinformation, image repurposing, also known as out-of-context misinformation, remains highly prevalent and effective. However, current approaches for detecting out-of-context misinformation often lack interpretability and offer limited explanations. In this study, we propose a logic regularization approach for out-of-context detection called LOGRAN (LOGic Regularization for out-of-context ANalysis). The primary objective of LOGRAN is to decompose the out-of-context detection at the phrase level. By employing latent variables for phrase-level predictions, the final prediction of the image-caption pair can be aggregated using logical rules. The latent variables also provide an explanation for how the final result is derived, making this fine-grained detection method inherently explanatory. We evaluate the performance of LOGRAN on the NewsCLIPpings dataset, showcasing competitive overall results. Visualized examples also reveal faithful phrase-level predictions of out-of-context images, accompanied by explanations. This highlights the effectiveness of our approach in addressing out-of-context detection and enhancing interpretability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Rethinking Graph Masked Autoencoders through Alignment and Uniformity

arXiv

引用

arXiv 2024年

作者： Wang, Liang Tao, Xiang Liu, Qiang Wu, Shu Wang, Liang Center for Research on Intelligent Perception and Computing 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

Self-supervised learning on graphs can be bifurcated into contrastive and generative methods. Contrastive methods, also known as graph contrastive learning (GCL), have dominated graph self-supervised learning in the past few years, but the recent advent of graph masked autoencoder (GraphMAE) rekindles the momentum behind generative methods. Despite the empirical success of GraphMAE, there is still a dearth of theoretical understanding regarding its efficacy. Moreover, while both generative and contrastive methods have been shown to be effective, their connections and differences have yet to be thoroughly investigated. Therefore, we theoretically build a bridge between GraphMAE and GCL, and prove that the node-level reconstruction objective in GraphMAE implicitly performs context-level GCL. Based on our theoretical analysis, we further identify the limitations of the GraphMAE from the perspectives of alignment and uniformity, which have been considered as two key properties of high-quality representations in GCL. We point out that GraphMAE's alignment performance is restricted by the masking strategy, and the uniformity is not strictly guaranteed. To remedy the aforementioned limitations, we propose an Alignment- Uniformity enhanced Graph Masked AutoEncoder, named AUG-MAE. Specifically, we propose an easy-to-hard adversarial masking strategy to provide hard-to-align samples, which improves the alignment performance. Meanwhile, we introduce an explicit uniformity regularizer to ensure the uniformity of the learned representations. Experimental results on benchmark datasets demonstrate the superiority of our model over existing state-of-the-art methods. The code is available at: https://***/AzureLeon1/AUG-MAE. Copyright © 2024, The Authors. All rights reserved.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Can Large Language Models Detect Rumors on Social Media?

arXiv

引用

arXiv 2024年

作者： Liu, Qiang Tao, Xiang Wu, Junfei Wu, Shu Wang, Liang Center for Research on Intelligent Perception and Computing 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

In this work, we investigate to use Large Language Models (LLMs) for rumor detection on social media. However, it is challenging for LLMs to reason over the entire propagation information on social media, which contains news contents and numerous comments, due to LLMs may not concentrate on key clues in the complex propagation information, and have trouble in reasoning when facing massive and redundant information. Accordingly, we propose an LLM-empowered Rumor Detection (LeRuD) approach, in which we design prompts to teach LLMs to reason over important clues in news and comments, and divide the entire propagation information into a Chain-of-Propagation for reducing LLMs’ burden. We conduct extensive experiments on the Twitter and Weibo datasets, and LeRuD outperforms several state-of-the-art rumor detection models by 3.2% to 7.7%. Meanwhile, by applying LLMs, LeRuD requires no data for training, and thus shows more promising rumor detection ability in few-shot or zero-shot scenarios. © 2024, CC BY.

关键词： Social networking (online)

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：