检索结果-内蒙古大学图书馆

Communications in Nonlinear science and Numerical Simulation 2022年 104卷

作者： Wei-Peng Nie Zhi-Dan Zhao Shi-Min Cai Tao Zhou Complex Lab School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu 610054 China Institute of Fundamental and Frontier Sciences University of Electronic Science and Technology of China Chengdu 610054 China Big Data Research Center University of Electronic Science and Technology of China Chengdu 610054 China Complexity Computation Laboratory Department of Computer Science School of Engineering Shantou University Shantou 515063 China Key Laboratory of Intelligent Manufacturing Technology (Ministry of Education) Shantou University Shantou 515063 China

Multimodal Collaboration networks for Geospatial Vehicle Detection in Dense, Occluded, and Large-Scale Events

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Xin Huang, Zhanchao Wang, Li Chanussot, Jocelyn Tian, Jiaojiao The School of Computer Science National Pilot Software Engineering School Beijing University of Posts and Telecommunications Beijing100876 China The Key Laboratory of Spatial Data Mining and Information Sharing The Ministry of Education The Academy of Digital China Fuzhou University Fuzhou350108 China The National & Local Joint Engineering Research Center of Satellite Geospatial Information Technology Fuzhou University Fuzhou350108 China The Univ. Grenoble Alpes CNRS Grenoble INP GIPSA-Lab Grenoble38000 France The Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China The Remote Sensing Technology Institute German Aerospace Center Weßling82205 Germany

In large-scale disaster events, the planning of optimal rescue routes depends on the object detection ability at the disaster scene, with one of the main challenges being the presence of dense and occluded objects. Existing methods, which are typically based on the RGB modality, struggle to distinguish targets with similar colors and textures in crowded environments and are unable to identify obscured objects. To this end, we first construct two multimodal dense and occlusion vehicle detection datasets for large-scale events, utilizing RGB and height map modalities. Based on these datasets, we propose a multimodal collaboration network for dense and occluded vehicle detection, MuDet for short. MuDet hierarchically enhances the completeness of discriminable information within and across modalities and differentiates between simple and complex samples. MuDet includes three main modules: Unimodal Feature Hierarchical Enhancement (Uni-Enh), Multimodal Cross Learning (Mul-Lea), and Hard-easy Discriminative (He-Dis) Pattern. Uni-Enh and Mul-Lea enhance the features within each modality and facilitate the cross-integration of features from two heterogeneous modalities. He-Dis effectively separates densely occluded vehicle targets with significant intra-class differences and minimal inter-class differences by defining and thresholding confidence values, thereby suppressing the complex background. Experimental results on two re-labeled multimodal benchmark datasets, the 4K-SAI-LCS dataset, and the ISPRS Potsdam dataset, demonstrate the robustness and generalization of the MuDet. The codes of this work are available openly at https://***/Shank2358/MuDet. © 2024, CC BY-NC-SA.

关键词： Remote sensing

Eocene Hyperthermal Events Drove Episodes of Vegetation Turnover in the Fushun Basin, Northeast China: Evidence from a Palaeoclimate Analysis of Palynological Assemblages

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Li, Yuanji Sun, Pingchang Falcon-Lang, Howard J. Liu, Zhaojun Zhang, Baoyong Zhang, Qiang Wang, Junxian Xu, Yinbo Department of Safety Engineering Heilongjiang University of Science and Technology Harbin150022 China National Professional Center Lab of Safety Basic Research for Hydrocarbon Gas Pipeline Transportation Network Harbin150022 China College of Earth Sciences Jilin University Jilin Changchun130061 China Key-Lab for Oil Shale and Paragenetic Minerals of Jilin Province Jilin Changchun130061 China Department of Earth Sciences Royal Holloway University of London Surrey EghamTW20 0EX United Kingdom Beijing Research Institute of Uranium Geology Beijing100029 China Oil and Gas Survey China Geological Survey Beijing100083 China

The response of terrestrial ecosystems to Eocene hyperthermal events remains relatively poorly understood. In this paper, we report palynological assemblages from an astronomically calibrated Eocene succession in the Fushun Basin of northeast China, in which previous δ 13 C studies have identified the Eocene Thermal Maxima 2 and 3 (ETM2, ETM3), the Early Eocene Climatic Optimum (EECO), and the Middle Eocene Climatic Optimum (MECO). In the samples that are directly associated with hyperthermal events, palynoassemblages show significant increases in the abundance of conifer pollen, especially Abietineaepollenites and Taxodiaceaepollenites , and fern spores of Polypodiaceaesporites , while angiosperm taxa reduce in frequency. Quantitative palaeoclimate parameters derived through the application of the Coexistence Approach to palynological assemblages indicate marked changes in temperature and humidity during hyperthermal events. These changes were most significant during ETM2 when mean annual temperature increased by 3–5℃ and mean annual precipitation increased by approximately 600 mm above background levels, but similar perturbations are also associated with the other hyperthermal events. An increase in organic abundance in lacustrine oil shale facies during hyperthermal events may imply that elevated humidity drove enhanced runoff, nutrient supply, and eutrophication of the hydrological open lake basins at a time when terrestrial carbon export from humid forests was also enhanced. Findings demonstrate that short-term bursts of Eocene warming drove major turnovers in the mid-latitude vegetation of Asia, reworking landscapes, and transforming regional hydrology. © 2022, The Authors. All rights reserved.

关键词： Vegetation

Tensor Decompositions for Hyperspectral data Processing in Remote Sensing: A Comprehensive Review

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Minghua Hong, Danfeng Han, Zhu Li, Jiaxin Yao, Jing Gao, Lianru Zhang, Bing Chanussot, Jocelyn The Key Laboratory of Computational Optical Imaging Technology Aerospace Information Research Institute Chinese Academy of Sciences Beijing10094 China The College of Resources and Environment University of Chinese Academy of Sciences Beijing100049 China The Key Laboratory of Digital Earth Science Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China The International Research Center of Big Data for Sustainable Development Goals Beijing100094 China The Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China The Univ. Grenoble Alpes CNRS Grenoble INP GIPSA-Lab Grenoble38000 France The Aerospace Information Research Institute Chinese Academy of Sciences Beijing10094 China

Owing to the rapid development of sensor technology, hyperspectral (HS) remote sensing (RS) imaging has provided a significant amount of spatial and spectral information for the observation and analysis of the Earth's surface at a distance of data acquisition devices, such as aircraft, spacecraft, and satellite. The recent advancement and even revolution of the HS RS technique offer opportunities to realize the full potential of various applications, while confronting new challenges for efficiently processing and analyzing the enormous HS acquisition data. Due to the maintenance of the 3-D HS inherent structure, tensor decomposition has aroused widespread concern and research in HS data processing tasks over the past decades. In this article, we aim at presenting a comprehensive overview of tensor decomposition, specifically contextualizing the five broad topics in HS data processing, and they are HS restoration, compressed sensing, anomaly detection, superresolution, and spectral unmixing. For each topic, we elaborate on the remarkable achievements of tensor decomposition models for HS RS with a pivotal description of the existing methodologies and a representative exhibition on the experimental results. As a result, the remaining challenges of the follow-up research directions are outlined and discussed from the perspective of the real HS RS practices and tensor decomposition merged with advanced priors and even with deep neural networks. This article summarizes different tensor decomposition-based HS data processing methods and categorizes them into different classes from simple adoptions to complex combinations with other priors for the algorithm beginners. We also expect this survey can provide new investigations and development trends for the experienced researchers who understand tensor decomposition and HS RS to some extent. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

Neuroimaging epicenters as potential sites of onset of the neuroanatomical pathology in schizophrenia

学校读者我要写书评

暂无评论

science Advances 2024年第24期10卷 eadk6063页

作者： Jiang, Yuchao Palaniyappan, Lena Luo, Cheng Chang, Xiao Zhang, Jie Tang, Yingying Zhang, Tianhong Li, Chunbo Zhou, Enpeng Yu, Xin Li, Wei An, Dongmei Zhou, Dong Huang, Chu-Chung Tsai, Shih-Jen Lin, Ching-Po Cheng, Jingliang Wang, Jijun Yao, Dezhong Cheng, Wei Feng, Jianfeng Institute of Science and Technology for Brain-Inspired Intelligence Fudan University Shanghai China Ministry of Education Shanghai China Douglas Mental Health University Institute Department of Psychiatry McGill University MontréalQC Canada Robarts Research Institute University of Western Ontario LondonON Canada Lawson Health Research Institute LondonON Canada Clinical Hospital of Chengdu Brain Science Institute MOE Key Lab for Neuroinformation School of life Science and Technology University of Electronic Science and Technology of China Chengdu China High-Field Magnetic Resonance Brain Imaging Key Laboratory of Sichuan Province Center for Information in Medicine University of Electronic Science and Technology of China Chengdu China Chinese Academy of Medical Sciences Chengdu China Shanghai Key Laboratory of Psychotic Disorders Shanghai Mental Health Center Shanghai Jiao Tong University School of Medicine Shanghai200030 China Beijing China Department of Neurology West China Hospital Sichuan University Chengdu610041 China School of Psychology and Cognitive Science East China Normal University Shanghai China Shanghai Changning Mental Health Center Shanghai China Department of Psychiatry Taipei Veterans General Hospital Taipei Taiwan Institute of Neuroscience National Yang Ming Chiao Tung University Taipei Taiwan Department of MRI First Affiliated Hospital of Zhengzhou University Zhengzhou China Department of Neurology Huashan Hospital Fudan University Shanghai China Fudan ISTBI—ZJNU Algorithm Centre for Brain-Inspired Intelligence Zhejiang Normal University Jinhua China MOE Frontiers Center for Brain Science Fudan University Shanghai China Zhangjiang Fudan International Innovation Center Shanghai China School of Data Science Fudan University Shanghai China Department of Computer Science University of Warwick CoventryCV 4 7AL United Kingdom

Schizophrenia lacks a clear definition at the neuroanatomical level, capturing the sites of origin and progress of this disorder. Using a network-theory approach called epicenter mapping on cross-sectional magnetic resonance imaging from 1124 individuals with schizophrenia, we identified the most likely "source of origin" of the structural pathology. Our results suggest that the Broca’s area and adjacent frontoinsular cortex may be the epicenters of neuroanatomical pathophysiology in schizophrenia. These epicenters can predict an individual’s response to treatment for psychosis. In addition, cross-diagnostic similarities based on epicenter mapping over of 4000 individuals diagnosed with neurological, neurodevelopmental, or psychiatric disorders appear to be limited. When present, these similarities are restricted to bipolar disorder, major depressive disorder, and obsessive-compulsive disorder. We provide a comprehensive framework linking schizophrenia-specific epicenters to multiple levels of neurobiology, including cognitive processes, neurotransmitter receptors and transporters, and human brain gene expression. Epicenter mapping may be a reliable tool for identifying the potential onset sites of neural pathophysiology in schizophrenia. © 2024 the Authors, some rights reserved

关键词： Mapping

Associations of social determinants of health with life expectancy and future health risks among individuals with type 2 diabetes: two nationwide cohort studies in the UK and USA

学校读者我要写书评

暂无评论

The Lancet Healthy Longevity 2024年第8期5卷 e542-e551页

作者： Zhong, Jiale Zhang, Yanbo Zhu, Kai Li, Rui Zhou, Xiaotao Yao, Pang Franco, Oscar H Manson, JoAnn E Pan, An Liu, Gang Department of Nutrition and Food Hygiene Hubei Key Laboratory of Food Nutrition and Safety Ministry of Education Key Lab of Environment and Health State Key Laboratory of Environment Health (Incubating) School of Public Health Tongji Medical College Huazhong University of Science and Technology Department of Epidemiology and Population Health Albert Einstein College of Medicine Department of Epidemiology and Biostatistics Ministry of Education Key Laboratory of Environment and Health School of Public Health Tongji Medical College Huazhong University of Science and Technology Public Health Service Center of Bao'an District Clinical Trial Service Unit and Epidemiological Studies Unit Nuffield Department of Population Health University of Oxford Department of Global Public Health & Bioethics University Medical Center Utrecht Department of Epidemiology Harvard T H Chan School of Public Health Channing Division of Network Medicine Department of Medicine Brigham and Women's Hospital and Harvard Medical School Division of Preventive Medicine Department of Medicine Brigham and Women's Hospital and Harvard Medical School

Background: Social determinants of health (SDHs) are the primary drivers of preventable health inequities, and the associations between SDHs and health outcomes among individuals with type 2 diabetes remain unclear. This study aimed to estimate the associations of combined SDHs with life expectancy and future health risks among adults with type 2 diabetes from the UK and USA. Methods: In an analysis of two nationwide cohort studies, adults with type 2 diabetes were identified from the UK Biobank from March 13, 2006, to Oct 1, 2010 (adults aged 37–73 years) and the US National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018 (adults aged ≥20 years). Participants with type 2 diabetes at baseline were included in our analysis. Participants without information on SDHs or follow-up were excluded. The UK Biobank assessed 17 SDHs and the US NHANES assessed ten SDHs, with each SDH dichotomised into advantaged and disadvantaged levels. The combined score of SDHs were calculated as the sum of the weighted scores for each SDH. Participants were then categorised into tertiles (favourable, medium, and unfavourable SDH groups). Primary outcomes were life expectancy and mortality in both cohorts, and incidences of cardiovascular disease, diabetes-related microvascular disease, dementia, and cancer in the UK Biobank. Outcomes were obtained from disease registries up until Dec 31, 2021, in the UK Biobank and Dec 31, 2019, in the US NHANES cohorts. Findings: We included 17 321 participants from the UK Biobank cohort (median age 61·0 years [IQR 56·0–65·0];6028 [34·8%] women and 11 293 [65·2%] men) and 7885 participants from the NHANES cohort (mean age 59·2 years [95% CI 58·7–59·6];3835 [49·1%, weighted] women and 4050 [50·9%, weighted] men) in our analysis. In the UK Biobank, 3235 deaths (median follow-up 12·3 years [IQR 11·5–13·2]), 3010 incident cardiovascular disease (12·1 years [10·8–13·0]), 1997 diabetes-related microvascular disease (8·0 years [7·1–8·9]), 773 d

关键词：

DRAC: Diabetic Retinopathy Analysis Challenge with Ultra-Wide Optical Coherence Tomography Angiography Images

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Qian, Bo Chen, Hao Wang, Xiangning Che, Haoxuan Kwon, Gitaek Kim, Jaeyoung Choi, Sungjin Shin, Seoyoung Krause, Felix Unterdechler, Markus Hou, Junlin Feng, Rui Li, Yihao Daho, Mostafa El Habib Wu, Qiang Zhang, Ping Yang, Xiaokang Cai, Yiyu Li, Huating Jia, Weiping Sheng, Bin Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong Department of Chemical and Biological Engineering The Hong Kong University of Science and Technology Hong Kong Department of Ophthalmology Shanghai Jiao Tong University Affiliated Sixth People’s Hospital Shanghai China VUNO Inc. Seoul Korea Republic of AI/DX Convergence Business Group KT Korea Republic of Johannes Kepler University Linz Linz Austria School of Computer Science Shanghai Key Laboratory of Intelligent Information Processing Fudan University China Academy for Engineering and Technology Fudan University China LaTIM UMR 1101 Inserm Brest France University of Western Brittany Brest France Department of Computer Science and Engineering The Ohio State University Ohio United States Department of Biomedical Informatics The Ohio State University Ohio United States Translational Data Analytics Institute The Ohio State University Ohio United States MoE Key Lab of Artificial Intelligence Artificial Intelligence Institute Shanghai Jiao Tong University Shanghai China School of Mechanical and Aerospace Engineering Nanyang Technological University Singapore Department of Endocrinology and Metabolism Shanghai Jiao Tong University Affiliated Sixth People’s Hospital Shanghai China Shanghai Diabetes Institute Shanghai Clinical Center for Diabetes Shanghai China

Computer-assisted automatic analysis of diabetic retinopathy (DR) is of great importance in reducing the risks of vision loss and even blindness. Ultra-wide optical coherence tomography angiography (UW-OCTA) is a non-invasive and safe imaging modality in DR diagnosis system, but there is a lack of publicly available benchmarks for model development and evaluation. To promote further research and scientific benchmarking for diabetic retinopathy analysis using UW-OCTA images, we organized a challenge named "DRAC - Diabetic Retinopathy Analysis Challenge" in conjunction with the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2022). The challenge consists of three tasks: segmentation of DR lesions, image quality assessment and DR grading. The scientific community responded positively to the challenge, with 11, 12, and 13 teams from geographically diverse institutes submitting different solutions in these three tasks, respectively. This paper presents a summary and analysis of the top-performing solutions and results for each task of the challenge. The obtained results from top algorithms indicate the importance of data augmentation, model architecture and ensemble of networks in improving the performance of deep learning models. These findings have the potential to enable new developments in diabetic retinopathy analysis. The challenge remains open for post-challenge registrations and submissions for benchmarking future methodology developments. Copyright © 2023, The Authors. All rights reserved.

关键词： Optical tomography

Incoherent feedforward loop dominates the robustness and tunability of necroptosis biphasic, emergent, and coexistent dynamics

学校读者我要写书评

暂无评论

Fundamental Research 2024年

作者： Xu, Fei Li, Xiang Wu, Rui Qi, Hong Jin, Jun Liu, Zhilong Wu, Yuning Lin, Hai Shen, Chuansheng Shuai, Jianwei Department of Physics and Fujian Provincial Key Laboratory for Soft Functional Materials Research Xiamen University Xiamen 361005 China National Institute for Data Science in Health and Medicine and State Key Laboratory of Cellular Stress Biology Innovation Center for Cell Signaling Network School of Life Sciences Xiamen University Xiamen 361102 China Complex Systems Research Center Shanxi University Taiyuan 030006 China Department of Mathematics and Physics Fujian Jiangxia University Fuzhou 350108 China Oujiang Laboratory (Zhejiang Lab for Regenerative Medicine Vision and Brain Health) and Wenzhou Institute University of Chinese Academy of Sciences Wenzhou 325001 China School of Mathematics and Physics Anqing Normal University Anqing 246011 China Department of Physics Anhui Normal University Wuhu 241002 China

Biphasic dynamics, the variable-dependent ability to enhance or restrain biological function, is prevalent in natural systems. Accompanied by biphasic dynamics, necroptosis signaling also appears emergent and coexistent dynamics. However, it remains elusive how the properties of these dynamics are characterized by specific circuit structures and components. Starting with necroptosis circuit modeling, we systematically analyzed the network topology for achieving RIP1-dependent biphasic, emergent, and coexistent (BEC) dynamics. RIP1-RIP3-Caspase-8 (C8) incoherent feedforward loop embedded with positive feedback of RIP3 to RIP1 is identified as the core topology. The peak value of RIP3 phosphorylation is determined to present a scale-invariant feature, dictating BEC dynamics and the bell-shaped regulation of necroptosis biphasic dynamics. To quantitatively determine the uncertainty of necroptosis coexistent dynamics, potential landscape and Shannon entropy that measure entropy production during cell death are introduced for the first time. Further random necroptosis circuit analysis identifies the bell-shaped regulation of necroptosis biphasic dynamics by RIP3 auto-phosphorylation, which acts as a complementary process for robustly attaining BEC dynamics. Finally, we searched all possible two- and three-node circuit topologies to screen those that could perform BEC dynamics. A complete atlas of three-node circuit BEC dynamics is generated and only three minimal circuits emerge as robust solutions, confirming incoherent feedforward loop is the core topology. Analysis of the association between the minimal circuit structure and robustness proves that the identified optimal functional achievement structure is highly consistent with the experimental observed RIP1-RIP3-C8 topology. Overall, through highlighting a finite set of circuits, this study yields guiding principles that enable the mapping, modulation, and design of circuits for BEC dynamics in diverse synthetic biol

关键词： Biphasic dynamics Design principle Emergence Incoherent feedforward loop Necroptosis

ChaLearn looking at people: IsoGD and ConGD large-scale RGB-D gesture recognition

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Wan, Jun Lin, Chi Wen, Longyin Li, Yunan Miao, Qiguang Escalera, Sergio Anbarjafari, Gholamreza Guyon, Isabelle Guo, Guodong Li, Stan Z. National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing100190 China JD Finance Mountain ViewCA United States University of Southern California Los AngelesCA90089-0911 United States School of Computer Science and Technology Xidian University & Xi'an Key Laboratory of Big Data and Intelligent Vision 2nd South Taibai Road Xi'an710071 China Universitat de Barcelona Computer Vision Center Spain iCV Lab Institute of Technology University of Tartu Estonia Faculty of Engineering Hasan Kalyoncu University Gaziantep Turkey Institute of Digital Technologies Loughborough University London United Kingdom ChaLearn United States University Paris-Saclay France institute of Deep Learning Baidu Research National Engineering Laboratory for Deep Learning Technology and Application China

The ChaLearn large-scale gesture recognition challenge has been run twice in two workshops in conjunction with the International Conference on Pattern Recognition (ICPR) 2016 and International Conference on Computer Vision (ICCV) 2017, attracting more than 200 teams around the world. This challenge has two tracks, focusing on isolated and continuous gesture recognition, respectively. This paper describes the creation of both benchmark datasets and analyzes the advances in large-scale gesture recognition based on these two datasets. We discuss the challenges of collecting large-scale ground-truth annotations of gesture recognition, and provide a detailed analysis of the current state-of-the-art methods for large-scale isolated and continuous gesture recognition based on RGB-D video sequences. In addition to recognition rate and mean jaccard index (MJI) as evaluation metrics used in our previous challenges, we also introduce the corrected segmentation rate (CSR) metric to evaluate the performance of temporal segmentation for continuous gesture recognition. Furthermore, we propose a bidirectional long short-term memory (Bi-LSTM) baseline method, determining the video division points based on the skeleton points extracted by convolutional pose machine (CPM). Experiments demonstrate that the proposed Bi-LSTM outperforms the state-of-the-art methods with an absolute improvement of 8.1% (from 0.8917 to 0.9639) of CSR. Copyright © 2019, The Authors. All rights reserved.

关键词： Long short-term memory

Active Learning for Name Entity Recognition with External Knowledge

学校读者我要写书评

暂无评论

ACM Transactions on Asian and Low-Resource Language Information Processing 1000年

作者： Ying Ma Yu Zhang Arun Kumar Sangaiah Ming Yan Guoqi Li Tian Wang Harbin Institute of Technology Xiamen University of Technology China Xiamen University of Technology China National Yunlin University of Science and Technology China Institute of High Performance Computing Agency for Science Technology and Research Singapore Institute of Automation China Academy of Sciences China BNU-UIC Institute of Artificial Intelligence and Future Networks Beijing Normal University Guangdong Key Lab of AI and Multi-Modal Data Processing BNU-HKBU United International College (UIC) China

Named Entity Recognition (NER) is an important task in knowledge extraction, which targets extracting structural information from unstructured text. To fully employ the prior-knowledge of the pre-trained language models, some research works formulate the NER task into the machine reading comprehension form (MRC-form) to enhance their model generalization capability of commonsense knowledge. However, this transformation still faces the data-hungry issue with limited training data for the specific NER tasks. To address the low-resource issue in NER, we introduce a method named active multi-task-based NER (AMT-NER), which is a two-stage multi-task active learning training model. Specifically, A multi-task learning module is first introduced into AMT-NER to improve its representation capability in low-resource NER tasks. Then, a two-stage training strategy is proposed to optimize AMT-NER multi-task learning. An associated task of Natural Language Inference (NLI) is also employed to enhance its commonsense knowledge further. More importantly, AMT-NER introduces an active learning module, uncertainty selective, to actively filter training data to help the NER model learn efficiently. Besides, we also find different external supportive data under different pipelines improves model performance differently in the NER tasks. Extensive experiments are performed to show the superiority of our method, which also proves our findings that the introduction of external knowledge is significant and effective in the MRC-form NER tasks.

关键词： Named Entity Recognition Machine Reading Comprehension Multi-task Learning Active Learning