检索结果-内蒙古大学图书馆

Neural quality estimation with multiple hypotheses for grammatical error correction

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Zhenghao Yi, Xiaoyuan Sun, Maosong Yang, Liner Chua, Tat-Seng Department of Computer Science and Technology Tsinghua University Beijing China Institute for Artificial Intelligence Tsinghua University Beijing China Beijing National Research Center for Information Science and Technology State Key Lab on Intelligent Technology and Systems Tsinghua University Beijing China Beijing Academy of Artificial Intelligence Beijing Language and Culture University Beijing China School of Computing National University of Singapore Singapore

Grammatical Error Correction (GEC) aims to correct writing errors and help language learners improve their writing skills. However, existing GEC models tend to produce spurious corrections or fail to detect lots of errors. The quality estimation model is necessary to ensure learners get accurate GEC results and avoid misleading from poorly corrected sentences. Well-trained GEC models can generate several high-quality hypotheses through decoding, such as beam search, which provide valuable GEC evidence and can be used to evaluate GEC quality. However, existing models neglect the possible GEC evidence from different hypotheses. This paper presents the Neural Verification Network (VERNet) for GEC quality estimation with multiple hypotheses. VERNet establishes interactions among hypotheses with a reasoning graph and conducts two kinds of attention mechanisms to propagate GEC evidence to verify the quality of generated hypotheses. Our experiments on four GEC datasets show that VERNet achieves state-of-the-art grammatical error detection performance, achieves the best quality estimation results, and significantly improves GEC performance by reranking hypotheses. All data and source codes are available at https://***/thunlp/VERNet. Copyright © 2021, The Authors. All rights reserved.

关键词： Error correction

Large language models for diabetes training:a prospective study

学校读者我要写书评

暂无评论

Science Bulletin 2025年第6期70卷 934-942页

作者： Haoxuan Li Zehua Jiang Zhouyu Guan Yuqian Bao Yuexing Liu Tingting Hu Jiajia Li Ruhan Liu Liang Wu Di Cheng Hongwei Ji Yong Wang Ya-Xing Wang Carol Y.Cheung Yingfeng Zheng Jihong Wang Zhen Li Weibing Wu Cynthia Ciwei Lim Yong Mong Bee Hong Chang Tan Elif I.Ekinci David C.Klonoff Justin B.Echouffo-Tcheugui Nestoras Mathioudakis Leonor Corsino Rafael Simó Charumathi Sabanayagam Gavin Siew Wei Tan Ching-Yu Cheng Tien Yin Wong Huating Li Chun Cai Lijuan Mao Lee-Ling Lim Yih-Chung Tham Bin Sheng Weiping Jia Shanghai University of Sport Shanghai 200438China School of Clinical Medicine Beijing Tsinghua Changgung HospitalTsinghua MedicineTsinghua UniversityBeijing 100084China Beijing Visual Science and Translational Eye Research Institute(BERI) Tsinghua MedicineTsinghua UniversityBeijing 100084China Shanghai Belt and Road International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases Department of Computer Science and EngineeringSchool of ElectronicInformationand Electrical EngineeringShanghai Jiao Tong UniversityDepartment of Endocrinology and MetabolismShanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of MedicineShanghai Diabetes InstituteShanghai Clinical Center for DiabetesShanghai 200240China MoE Key Lab of Artificial Intelligence Artificial Intelligence InstituteShanghai Jiao Tong UniversityShanghai 200240China Department of Cardiology the Affiliated Hospital of Qingdao UniversityQingdao 266011China CEMS NCMISHCMSMADISAcademy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing 100190China School of Mathematics University of Chinese Academy of SciencesChinese Academy of SciencesBeijing 100049China Key Laboratory of Systems Health Science of Zhejiang Province School of Life ScienceHangzhou Institute for Advanced StudyUniversity of Chinese Academy of SciencesHangzhou 310024China Beijing Institute of Ophthalmology Beijing Tongren HospitalCapital Medical UniversityBeijing Ophthalmology and Visual Sciences Key LaboratoryBeijing 100730China k.Department of Ophthalmology and Visual Sciences The Chinese University of Hong KongHong Kong 999077China State Key Laboratory of Ophthalmology Zhongshan Ophthalmic CenterSun Yat-sen UniversityGuangdong Provincial Key Laboratory of Ophthalmology and Visual ScienceGuangdong Provincial Clinical Research Center for Ocular DiseasesGuangzhou 510060China Department of Renal Medicine Singapore General HospitalSingHealth-Duke Academic Medical CentreSingapore169

Diabetes poses a considerable global health challenge,with varying levels of diabetes knowledge among healthcare professionals,highlighting the importance of diabetes *** Language Models(LLMs)provide new insights into diabetes training,but their performance in diabetes-related queries remains uncertain,especially outside the English language like *** first evaluated the performance of ten LLMs:ChatGPT-3.5,ChatGPT-4.0,Google Bard,LlaMA-7B,LlaMA2-7B,Baidu ERNIE Bot,Ali Tongyi Qianwen,MedGPT,HuatuoGPT,and Chinese LlaMA2-7B on diabetes-related queries,based on the Chinese National Certificate Examination for Primary Diabetes Care in China(NCE-CPDC)and the English Specialty Certificate Examination in Endocrinology and Diabetes of Membership of the Royal College of Physicians of the United ***,we assessed the training of primary care physicians(PCPs)without and with the assistance of ChatGPT-4.0 in the NCE-CPDC examination to ascertain the reliability of LLMs as medical *** found that ChatGPT-4.0 outperformed other LLMs in the English examination,achieving a passing accuracy of 62.50%,which was significantly higher than that of Google Bard,LlaMA-7B,and *** the NCE-CPFC examination,ChatGPT-4.0,Ali Tongyi Qianwen,Baidu ERNIE Bot,Google Bard,MedGPT,and ChatGPT-3.5 successfully passed,whereas LlaMA2-7B,HuatuoGPT,Chinese LLaMA2-7B,and LlaMA-7B ***-4.0(84.82%)surpassed all PCPs and assisted most PCPs in the NCE-CPDC examination(improving by 1%–6.13%).In summary,LLMs demonstrated outstanding competence for diabetes-related questions in both the Chinese and English language,and hold great potential to assist future diabetes training for physicians globally.

关键词： Diabetes Diabetestraining Largelanguagemodels Primarydiabetescare Prospectivestudy

Stylized dialogue response generation using stylized unpaired texts

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zheng, Yinhe Chen, Zikai Zhang, Rongsheng Huang, Shilei Mao, Xiaoxi Hang, Minlie Department of Computer Science and Technology Institute for Artifical Intelligence State Key Lab of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Tsinghua University Beijing China Fuxi AI Lab NetEase Inc. Hangzhou China Beijing China

Generating stylized responses is essential to build intelligent and engaging dialogue systems. However, this task is far from well-explored due to the difficulties of rendering a particular style in coherent responses, especially when the target style is embedded only in unpaired texts that cannot be directly used to train the dialogue model. This paper proposes a stylized dialogue generation method that can capture stylistic features embedded in unpaired texts. Specifically, our method can produce dialogue responses that are both coherent to the given context and conform to the target style. In this study, an inverse dialogue model is first introduced to predict possible posts for the input responses, and then this inverse model is used to generate stylized pseudo dialogue pairs based on these stylized unpaired texts. Further, these pseudo pairs are employed to train the stylized dialogue model with a joint training process, and a style routing approach is proposed to intensify stylistic features in the decoder. Automatic and manual evaluations on two datasets demonstrate that our method outperforms competitive baselines in producing coherent and style-intensive dialogue responses. Copyright © 2020, The Authors. All rights reserved.

关键词： Speech processing

A self-training method for machine reading comprehension with soft evidence extraction

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Niu, Yilin Jiao, Fangkai Zhou, Mantong Yao, Ting Xu, Jingfang Huang, Minlie Department of Computer Science and Technology Institute for Artificial Intelligence State Key Lab of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China School of Computer Science and Technology Shandong University Sogou Inc. Beijing China

Neural models have achieved great success on machine reading comprehension (MRC), many of which typically consist of two components: an evidence extractor and an answer predictor. The former seeks the most relevant information from a reference text, while the latter is to locate or generate answers from the extracted evidence. Despite the importance of evidence labels for training the evidence extractor, they are not cheaply accessible, particularly in many non-extractive MRC tasks such as YES/NO question answering and multi-choice MRC. To address this problem, we present a Self-Training method (STM), which supervises the evidence extractor with auto-generated evidence labels in an iterative process. At each iteration, a base MRC model is trained with golden answers and noisy evidence labels. The trained model will predict pseudo evidence labels as extra supervision in the next iteration. We evaluate STM on seven datasets over three MRC tasks. Experimental results demonstrate the improvement on existing MRC models, and we also analyze how and why such a self-training method works in MRC. Copyright © 2020, The Authors. All rights reserved.

关键词： Iterative methods

Directly wireless communication of human minds via non-invasive brain-computer-metasurface platform

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ma, Qian Gao, Wei Xiao, Qiang Ding, Lingsong Gao, Tianyi Zhou, Yajun Gao, Xinxin Yan, Tao Liu, Che Gu, Ze Kong, Xianghong Abbasi, Qammer H. Li, Lianlin Qiu, Cheng-Wei Li, Yuanqing Cui, Tie Jun Institute of Electromagnetic Space Southeast University Nanjing210096 China School of Automation Science and Engineering South China University of Technology Guangzhou510641 China State Key Laboratory of Millimeter Wave Southeast University Nanjing210096 China Center of Intelligent Metamaterials Pazhou Laboratory Guangzhou510330 China Research Center for Brain-Computer Interface Pazhou Lab Guangzhou510330 China Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore University of Glasgow James Watt School of Engineering GlasgowG12 8QQ United Kingdom State Key Laboratory of Advanced Optical Communication Systems and Networks Department of Electronics Peking University Beijing100871 China

Brain-computer interfaces (BCIs), invasive or non-invasive, have projected unparalleled vision and promise for assisting patients in need to better their interaction with the surroundings. Inspired by the BCI-based rehabilitation technologies for nerve-system impairments and amputation, we propose an electromagnetic brain-computer-metasurface (EBCM) paradigm, regulated by human's cognition by brain signals directly and non-invasively. We experimentally show that our EBCM platform can translate human's mind from evoked potentials of P300-based electroencephalography to digital coding information in the electromagnetic domain non-invasively, which can be further processed and transported by an information metasurface in automated and wireless fashions. Directly wireless communications of the human minds are performed between two EBCM operators with accurate text transmissions. Moreover, several other proof-of-concept mind-control schemes are presented using the same EBCM platform, exhibiting flexibly-customized capabilities of information processing and synthesis like visual-beam scanning, wave modulations, and pattern encoding. © 2022, CC BY.

关键词： Electrophysiology

Reusing Discriminators for Encoding: Towards Unsupervised Image-to-Image Translation

学校读者我要写书评

暂无评论

Reusing Discriminators for Encoding: Towards Unsupervised Im...

Conference on computer vision and Pattern Recognition (CVPR)

作者： Runfa Chen Wenbing Huang Binghui Huang Fuchun Sun Bin Fang Department of Computer Science and Technology Tsinghua University Institute for Artificial Intelligence Tsinghua University (THUAI) Beijing National Research Center for Information Science and Technology (BNRist) State Key Lab on Intelligent Technology and Systems Beijing P.R.China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

Unsupervised image-to-image translation is a central task in computer vision. Current translation frameworks will abandon the discriminator once the training process is completed. This paper contends a novel role of the discriminator by reusing it for encoding the images of the target domain. The proposed architecture, termed as NICE-GAN, exhibits two advantageous patterns over previous approaches: First, it is more compact since no independent encoding component is required; Second, this plug-in encoder is directly trained by the adversary loss, making it more informative and trained more effectively if a multi-scale discriminator is applied. The main issue in NICE-GAN is the coupling of translation with discrimination along the encoder, which could incur training inconsistency when we play the min-max game via GAN. To tackle this issue, we develop a decoupled training strategy by which the encoder is only trained when maximizing the adversary loss while keeping frozen otherwise. Extensive experiments on four popular benchmarks demonstrate the superior performance of NICE-GAN over state-of-the-art methods in terms of FID, KID, and also human preference. Comprehensive ablation studies are also carried out to isolate the validity of each proposed component. Our codes are available at https://***/alpc91/NICE-GAN-pytorch.

关键词： Training Encoding Gallium nitride Generators Image coding Task analysis Benchmark testing

The Lagrangian remainder of Taylor’s series, distinguishes O(f(x)) time complexities to polynomials or not

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Bakas, Nikolaos P. Kosmatopoulos, Elias Chatzichristofis, Savvas A. Computation-based Science and Technology Research Center Cyprus Institute 20 Konstantinou Kavafi Street Aglantzia Nicosia2121 Cyprus Democritus University of Thrace Electrical and Computer engineering Intelligent Systems Lab Department of Computer Science Neapolis University Pafos 2 Danais Avenue Pafos8042 Cyprus

The purpose of this letter is to investigate the time complexity consequences of the truncated Taylor series, known as Taylor Polynomials [1–3]. In particular, it is demonstrated that the examination of the P = NP equality, is associated with the determination of whether the nth derivative of a particular solution is bounded or not. Accordingly, in some cases, this is not true, and hence in general. Copyright © 2020, The Authors. All rights reserved.

关键词： Taylor series

Human action recognition in drone videos using a few aerial training examples

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Sultani, Waqas Shah, Mubarak Intelligent Machine Lab Information Technology University Lahore Pakistan Center for Research in Computer Vision University of Central Florida Orlando United States

Drones are enabling new forms of human actions surveillance due to their low cost and fast mobility. However, using deep neural networks for automatic aerial action recognition is difficult due to the need for a large number of training aerial human action videos. Collecting a large number of human action aerial videos is costly, time-consuming, and difficult. In this paper, we explore two alternative data sources to improve aerial action classification when only a few training aerial examples are available. As a first data source, we resort to video games. We collect plenty of aerial game action videos using two gaming engines. For the second data source, we leverage conditional Wasserstein Generative Adversarial Networks to generate aerial features from ground videos. Given that both data sources have some limitations, e.g. game videos are biased towards specific actions categories (fighting, shooting, etc.,), and it is not easy to generate good discriminative GAN-generated features for all types of actions, we need to efficiently integrate two dataset sources with few available real aerial training videos. To address this challenge of the heterogeneous nature of the data, we propose to use a disjoint multitask learning framework. We feed the network with real and game, or real and GAN-generated data in an alternating fashion to obtain an improved action classifier. We validate the proposed approach on two aerial action datasets and demonstrate that features from aerial game videos and those generated from GAN can be extremely useful for an improved action recognition in real aerial videos when only a few real aerial training examples are available. Copyright © 2019, The Authors. All rights reserved.

关键词： Antennas

When counterpoint meets chinese folk melodies 20

学校读者我要写书评

暂无评论

When counterpoint meets chinese folk melodies

Proceedings of the 34th International Conference on Neural Information Processing systems

作者： Nan Jiang Sheng Jin Zhiyao Duan Changshui Zhang Institute for Artificial Intelligence Tsinghua University (THUAI) and State Key Lab of Intelligent Technologies and Systems and Beijing National Research Center for Information Science and Technology (BNRist) and Department of Automation Tsinghua University Beijing China Department of Electrical and Computer Engineering University of Rochester

ISBN: (纸本)9781713829546

Counterpoint is an important concept in Western music theory. In the past century, there have been significant interests in incorporating counterpoint into Chinese folk music composition. In this paper, we propose a reinforcement learning-based system, named FolkDuet, towards the online countermelody generation for Chinese folk melodies. With no existing data of Chinese folk duets, FolkDuet employs two reward models based on out-of-domain data i.e., Bach chorales, and monophonic Chinese folk melodies. An interaction reward model is trained on the duets formed from outer parts of Bach chorales to model counterpoint interaction, while a style reward model is trained on monophonic melodies of Chinese folk songs to model melodic patterns. With both rewards, the generator of FolkDuet is trained to generate countermelodies while maintaining the Chinese folk style. The entire generation process is performed in an online fashion, allowing real-time interactive human-machine duet improvisation. Experiments show that the proposed algorithm achieves better subjective and objective results than the baselines.

关键词：