检索结果-内蒙古大学图书馆

Adaptive Loose Optimization for Robust Question Answering

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ma, Jie Wang, Pinghui Wang, Zewei Kong, Dechen Hu, Min Han, Ting Liu, Jun The Ministry of Education of Key Laboratory for Intelligent Networks and Network Security School of Cyber Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China The Ministry of Education of Key Laboratory for Intelligent Networks and Network Security School of Automation Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China The China Mobile Research Institute China The Shannxi Provincial Key Laboratory of Big Data Knowledge Engineering School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Xi’an710049 China

Question answering methods are well-known for leveraging data bias, such as the language prior in visual question answering and the position bias in machine reading comprehension (extractive question answering). Current debiasing methods often come at the cost of significant in-distribution performance to achieve favorable out-of-distribution generalization ability, while non-debiasing methods sacrifice a considerable amount of out-of-distribution performance in order to obtain high in-distribution performance. Therefore, it is challenging for them to deal with the complicated changing real-world situations. In this paper, we propose a simple yet effective novel loss function with adaptive loose optimization, which seeks to make the best of both worlds for question answering. Our main technical contribution is to reduce the loss adaptively according to the ratio between the previous and the current optimization state on mini-batch training data. This loose optimization can be used to prevent non-debiasing methods from overlearning data bias while enabling debiasing methods to maintain slight bias learning. Experiments on the visual question answering datasets, including VQA v2, VQA-CP v1, VQA-CP v2, GQA-OOD, and the extractive question answering dataset SQuAD demonstrate that our approach enables QA methods to obtain state-of-the-art in- and out-of-distribution performance in most cases. The source code has been released publicly in https://***/reml-group/ALO. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

ViGT: Proposal-free Video Grounding with Learnable Token in Transformer

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Kun Guo, Dan Wang, Meng School of Computer Science and Information Engineering Hefei University of Technology Hefei230601 China Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Hefei230601 China Intelligent Interconnected Systems Laboratory of Anhui Province Hefei230601 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230088 China

The video grounding (VG) task aims to locate the queried action or event in an untrimmed video based on rich linguistic descriptions. Existing proposal-free methods are trapped in complex interaction between video and query, overemphasizing cross-modal feature fusion and feature correlation for VG. In this paper, we propose a novel boundary regression paradigm that performs regression token learning in a transformer. Particularly, we present a simple but effective proposal-free framework, namely Video Grounding Transformer (ViGT), which predicts the temporal boundary using a learnable regression token rather than multi-modal or cross-modal features. In ViGT, the benefits of a learnable token are manifested as follows. (1) The token is unrelated to the video or the query and avoids data bias toward the original video and query. (2) The token simultaneously performs global context aggregation from video and query features. First, we employed a sharing feature encoder to project both video and query into a joint feature space before performing cross-modal co-attention (i.e., video-to-query attention and query-to-video attention) to highlight discriminative features in each modality. Furthermore, we concatenated a learnable regression token [REG] with the video and query features as the input of a vision-language transformer. Finally, we utilized the token [REG] to predict the target moment and visual features to constrain the foreground and background probabilities at each timestamp. The proposed ViGT performed well on three public datasets: ANet Captions, TACoS and YouCookII. Extensive ablation studies and qualitative analysis further validated the interpretability of ViGT. Copyright © 2023, The Authors. All rights reserved.

关键词： Regression analysis

GPTR: Gestalt-Perception Transformer for Diagram Object Detection

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Hu, Xin Zhang, Lingling Liu, Jun Fan, Jinfu You, Yang Wu, Yaqiang Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering School of Computer Science and Technology Xi’an Jiaotong University China National Engineering Lab for Big Data Analytics Xi’an Jiaotong University China Department of Control Science and Engineering Tongji University Shanghai China Department of Computer Science National University of Singapore Singapore Lenovo Research Beijing China

Diagram object detection is the key basis of practical applications such as textbook question answering. Because the diagram mainly consists of simple lines and color blocks, its visual features are sparser than those of natural images. In addition, diagrams usually express diverse knowledge, in which there are many low-frequency object categories in diagrams. These lead to the fact that traditional data-driven detection model is not suitable for diagrams. In this work, we propose a gestalt-perception transformer model for diagram object detection, which is based on an encoder-decoder architecture. Gestalt perception contains a series of laws to explain human perception, that the human visual system tends to perceive patches in an image that are similar, close or connected without abrupt directional changes as a perceptual whole object. Inspired by these thoughts, we build a gestalt-perception graph in transformer encoder, which is composed of diagram patches as nodes and the relationships between patches as edges. This graph aims to group these patches into objects via laws of similarity, proximity, and smoothness implied in these edges, so that the meaningful objects can be effectively detected. The experimental results demonstrate that the proposed GPTR achieves the best results in the diagram object detection task. Our model also obtains comparable results over the competitors in natural image object detection. Copyright © 2022, The Authors. All rights reserved.

关键词： Object detection

A Tutorial on Movable Antennas for Wireless Networks

学校读者我要写书评

暂无评论

IEEE Communications Surveys and Tutorials 2025年

作者： Zhu, Lipeng Ma, Wenyan Mei, Weidong Zeng, Yong Wu, Qingqing Ning, Boyu Xiao, Zhenyu Shao, Xiaodan Zhang, Jun Zhang, Rui National University of Singapore Department of Electrical and Computer Engineering 117583 Singapore National Key Laboratory of Wireless Communications Chengdu611731 China Southeast University National Mobile Communications Research Laboratory Frontiers Science Center for Mobile Information Communication and Security Nanjing210096 China Purple Mountain Laboratories Nanjing211111 China Shanghai Jiao Tong University Department of Electronic Engineering Shanghai200240 China Beihang University School of Electronic and Information Engineering Beijing100191 China University of Waterloo Department of Electrical and Computer Engineering WaterlooONN2L 3G1 Canada Beijing Institute of Technology State Key Laboratory of CNS/ATM Miit Key Laboratory of Complex-field Intelligent Sensing Beijing100081 China The Chinese University of HongKong School of Science and Engineering Shenzhen Research Institute of Big Data Guangdong Shenzhen518172 China

Movable antenna (MA) has been recognized as a promising technology to enhance the performance of wireless communication and sensing by enabling antenna movement. Such a significant paradigm shift from conventional fixed antennas (FAs) to MAs offers tremendous new opportunities towards realizing more versatile, adaptive and efficient next-generation wireless networks such as 6G. In this paper, we provide a comprehensive tutorial on the fundamentals and advancements in the area of MA-empowered wireless networks. First, we overview the historical development and contemporary applications of MA technologies. Next, to characterize the continuous variation in wireless channels with respect to antenna position and/or orientation, we present new field-response channel models tailored for MAs, which are applicable to narrowband and wideband systems as well as far-field and near-field propagation conditions. Subsequently, we review the state-of-the-art architectures for implementing MAs and discuss their practical constraints. A general optimization framework is then formulated to fully exploit the spatial degrees of freedom (DoFs) in antenna movement for performance enhancement in wireless systems. In particular, we delve into two major design issues for MA systems. First, we address the intricate antenna movement optimization problem for various communication and/or sensing systems to maximize the performance gains achievable by MAs. Second, we deal with the challenging channel acquisition issue in MA systems for reconstructing the channel mapping between arbitrary antenna positions inside the transmitter and receiver regions. Moreover, we show existing prototypes developed for MA-aided communication/sensing and the experimental results based on them. Finally, the extension of MA design to other wireless systems and its synergy with other emerging wireless technologies are discussed. We also highlight promising research directions in this area to inspire future investigatio

关键词： Antenna arrays

Image Fusion Method Based on NSCT and Adaptive Sparse Representation 4

学校读者我要写书评

暂无评论

Image Fusion Method Based on NSCT and Adaptive Sparse Repres...

4th International Conference on Video and Image Processing, ICVIP 2020

作者： Dong, Zhangyu Wei, Xin Wang, Mengyang School of Computer and Information Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Hefei230009 China

ISBN: (纸本)9781450389075

ABSTRACT: Aiming at the imaging characteristics of multispectral and panchromatic image and the problem that the traditional image fusion methods cannot take into account the spectral quality and spatial resolution of the image at the same time, a novel fusion method based on the Non-subsampled Contourlet Transform (NSCT) domain and adaptive sparse representation (ASR) algorithm is presented in this paper. It combines the advantages of principal component substitution and multi-scale analysis, and uses the ASR model to encode the low-frequency components, which improves the shortcomings of the traditional SR-based approach applications. Firstly, NSCT transform is performed on thefirst principal component of multi-spectral image obtained by PCA transform and panchromatic image,and the low-pass and high-pass coefficients of the source image are obtained. Secondly, the Max-absolute' rule and the ASR model are used to fuse the separated high-pass and low-pass bands respectively. Finally, we obtained the final fused image by the corresponding inverse transformation reconstruction. The results show that the fusion algorithm proposed in this paper performs well in both spatial resolution and spectral performance;it not only improves the subjective visual effect, but also improves the objective evaluation index. © 2020 ACM.

关键词： Image fusion

Scientific Value Weights more than Being Open or Toll Access:An analysis of the OA advantage in Nature and Science

学校读者我要写书评

暂无评论

Journal of data and Information Science 2021年第4期6卷 62-75页

作者： Howell Y.Wang Shelia X.Wei Cong Cao Xianwen Wang Fred Y.Ye International Joint Informatics Laboratory&Jiangsu Key Laboratory of Data Engineering and Knowledge Service School of Information ManagementNanjing UniversityNanjing 210023China Nottingham University Business School China University of Nottingham Ningbo ChinaNingbo315100China WISE Lab. Dalian University of TechnologyDalian 116024China

Purpose:We attempt to find out whether OA or TA really affects the dissemination of scientific ***/methodology/approach:We design the indicators,hot-degree,and R-index to indicate a topic OA or TA ***,according to the OA classification of the Web of Science(WoS),we collect data from the WoS by downloading OA and TA articles,letters,and reviews published in Nature and Science during 2010–*** papers are divided into three broad disciplines,namely biomedicine,physics,and ***,taking a discipline in a journal and using the classical Latent Dirichlet Allocation(LDA)to cluster 100 topics of OA and TA papers respectively,we apply the Pearson correlation coefficient to match the topics of OA and TA,and calculate the hot-degree and R-index of every OA-TA topic ***,characteristics of the discipline can be *** qualitative comparison,we choose some high-quality papers which belong to Nature remarkable papers or Science breakthroughs,and analyze the relations between OA/TA and citation ***:The result shows that OA hot-degree in biomedicine is significantly greater than that of TA,but significantly less than that of TA in *** on the R-index,it is found that OA advantages exist in biomedicine and TA advantages do in ***,the dissemination of average scientific discoveries in all fields is not necessarily affected by OA or ***,OA promotes the spread of important scientific discoveries in high-quality *** limitations:We lost some citations by ignoring other open sources such as arXiv and *** limitation came from that Nature employs some strong measures for access-promoting subscription-based articles,on which the boundary between OA and TA became *** implications:It is useful to select hot topics in a set of publications by the hotdegree *** finding comprehensively reflects the differences of OA and TA in different disciplines,which is a u

关键词： Open access Toll access Scientific discovery Academic dissemination data analysis

Extremely Short Chinese Text Classification Method Based on Bidirectional Semantic Extension 2

学校读者我要写书评

暂无评论

Extremely Short Chinese Text Classification Method Based on ...

2019 2nd International Symposium on Big data and Applied Statistics, ISBDAS 2019

作者： Yue, Yongzeng Zhang, Yuhong Hu, Xuegang Li, Peipei Key Laboratory of Knowledge Engineering with Big Data [Hefei University of Technology] Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Hefei23061 China

Short text classification methods have achieved significant progress and wide application on text data such as Twitter and Weibo. However, the extremely short chinese texts like tax invoice data are different with traditional short texts in lackness of contextual semantic information, feature sparseness and extremely short length. The existing short text classification methods are difficult to achieve a satisfactory performance in these texts. To address these problems, this paper proposes a text classification method based on bidirectional semantic extension for extremely short texts like Chinese tax invoice data. More specifically, firstly, the Chinese knowledge graph is introduced for extending bidirectional semantic of texts and label data to expand the extremely short texts and ease the problem of feature sparseness;secondly, the hash vectorization is used to avoid the semantic problem caused by the lackness of contextual information. Experimental results conducted the real tax invoice dataset demonstrate the effectiveness of our proposed method. © Published under licence by IOP Publishing Ltd.

关键词： Classification (of information)

Semi-autoregressive transformer for image captioning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhou, Yuanen Zhang, Yong Hu, Zhenzhen Wang, Meng School of Computer Science and Information Engineering Hefei University of Technology Tencent AI Lab Key Laboratory of Knowledge Engineering with Big Data Ministry of Education

Current state-of-the-art image captioning models adopt autoregressive decoders, i.e. they generate each word by conditioning on previously generated words, which leads to heavy latency during inference. To tackle this issue, non-autoregressive image captioning models have recently been proposed to significantly accelerate the speed of inference by generating all words in parallel. However, these non-autoregressive models inevitably suffer from large generation quality degradation since they remove words dependence excessively. To make a better trade-off between speed and quality, we introduce a semi-autoregressive model for image captioning (dubbed as SATIC), which keeps the autoregressive property in global but generates words parallelly in local. Based on Transformer, there are only a few modifications needed to implement SATIC. Experimental results on the MSCOCO image captioning benchmark show that SATIC can achieve a good trade-off without bells and whistles. Code is available at https://***/YuanEZhou/satic. Copyright © 2021, The Authors. All rights reserved.

关键词： Economic and social effects

Graph-based exercise- And knowledge-aware learning network for student performance prediction

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Mengfan Shao, Pengyang Zhang, Kun Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology China School of Computer Science and Information Engineering Hefei University of Technology China

Predicting student performance is a fundamental task in Intelligent Tutoring Systems (ITSs), by which we can learn about students' knowledge level and provide personalized teaching strategies for them. Researchers have made plenty of efforts on this task. They either leverage educational psychology methods to predict students' scores according to the learned knowledge proficiency, or make full use of Collaborative Filtering (CF) models to represent latent factors of students and exercises. However, most of these methods either neglect the exercise-specific characteristics (e.g., exercise materials), or cannot fully explore the high-order interactions between students, exercises, as well as knowledge concepts. To this end, we propose a Graph-based Exercise- and knowledge-Aware Learning Network for accurate student score prediction. Specifically, we learn students' mastery of exercises and knowledge concepts respectively to model the two-fold effects of exercises and knowledge concepts. Then, to model the high-order interactions, we apply graph convolution techniques in the prediction process. Extensive experiments on two real-world datasets prove the effectiveness of our proposed Graph-EKLN. Copyright © 2021, The Authors. All rights reserved.

关键词： Students