检索结果-内蒙古大学图书馆

A knowledge-guided framework for frame identification 59

学校读者我要写书评

暂无评论

A knowledge-guided framework for frame identification

Joint Conference of the 59th Annual Meeting of the Association for computational Linguistics and the 11th International Joint Conference on Natural Language processing, ACL-IJCNLP 2021

作者： Su, Xuefeng Li, Ru Li, Xiaoli Pan, Jeff Z. Zhang, Hu Chai, Qinghua Han, Xiaoqi School of Computer and Information Technology Shanxi University Taiyuan China School of E-Commerce and Logistics Shanxi Vocational University of Engineering Technology Taiyuan China Key Laboratory of Computational Intelligence Chinese Information Processing of Ministry of Education Shanxi University Taiyuan China Institute for Infocomm Research A*Star Singapore ILCC School of Informatics University of Edinburgh United Kingdom

ISBN: (纸本)9781954085527

Frame Identification (FI) is a fundamental and challenging task in frame semantic parsing. The task aims to find the exact frame evoked by a target word in a given sentence. It is generally regarded as a classification task in existing work, where frames are treated as discrete labels or represented using one-hot embeddings. However, the valuable knowledge about frames is neglected. In this paper, we propose a Knowledge-Guided Frame Identification framework (KGFI) that integrates three types frame knowledge, including frame definitions, frame elements and frame-to frame relations, to learn better frame representation, which guides the KGFI to jointly map target words and frames into the same embedding space and subsequently identify the best frame by calculating the dot-product similarity scores between the target word embedding and all of the frame embeddings. The extensive experimental results demonstrate KGFI significantly outperforms the state-of-the-art methods on two benchmark datasets. © 2021 Association for computational Linguistics

关键词： Embeddings

DMNER: Biomedical Named Entity Recognition by Detection and Matching

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Bian, Junyi Jiang, Rongze Zhai, Weiqi Huang, Tianyang Zhou, Hong Zhu, Shanfeng School of Computer Science Fudan University Shanghai200433 China Atypon Systems LLC United Kingdom Institute of Science and Technology for Brain-Inspired Intelligence Fudan University China Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence Fudan University Ministry of Education Shanghai200433 China MOE Frontiers Center for Brain Science Fudan University Shanghai200433 China Zhangjiang Fudan International Innovation Center Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai200433 China

Biomedical named entity recognition (BNER) serves as the foundation for numerous biomedical text mining tasks. Unlike general NER, BNER require a comprehensive grasp of the domain, and incorporating external knowledge beyond training data poses a significant challenge. In this study, we propose a novel BNER framework called DMNER. By leveraging existing entity representation models SAP-BERT, we tackle BNER as a two-step process: entity boundary detection and biomedical entity matching. DMNER exhibits applicability across multiple NER scenarios: 1) In supervised NER, we observe that DMNER effectively rectifies the output of baseline NER models, thereby further enhancing performance. 2) In distantly supervised NER, combining MRC and AutoNER as span boundary detectors enables DMNER to achieve satisfactory results. 3) For training NER by merging multiple datasets, we adopt a framework similar to DS-NER but additionally leverage ChatGPT to obtain high-quality phrases in the training. Through extensive experiments conducted on 10 benchmark datasets, we demonstrate the versatility and effectiveness of DMNER. Copyright © 2023, The Authors. All rights reserved.

关键词： Character recognition

RFN-Nest: An end-to-end residual fusion network for infrared and visible images

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Hui Wu, Xiao-Jun Kittler, Josef Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence School of Artificial Intelligence and Computer Science Jiangnan University Wuxi214122 China The Center for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

In the image fusion field, the design of deep learning-based fusion methods is far from routine. It is invariably fusion-task specific and requires a careful consideration. The most difficult part of the design is to choose an appropriate strategy to generate the fused image for a specific task in hand. Thus, devising learnable fusion strategy is a very challenging problem in the community of image fusion. To address this problem, a novel end-to-end fusion network architecture (RFN-Nest) is developed for infrared and visible image fusion. We propose a residual fusion network (RFN) which is based on a residual architecture to replace the traditional fusion approach. A novel detail-preserving loss function, and a feature enhancing loss function are proposed to train RFN. The fusion model learning is accomplished by a novel two-stage training strategy. In the first stage, we train an auto-encoder based on an innovative nest connection (Nest) concept. Next, the RFN is trained using the proposed loss functions. The experimental results on public domain data sets show that, compared with the existing methods, our end-to-end fusion network delivers a better performance than the state-of-the-art methods in both subjective and objective evaluation. The code of our fusion method is available at https://***/hli1221/imagefusion-rfn-nest. © 2021, CC BY-NC-ND.

关键词： Image fusion

PPT Fusion: Pyramid Patch Transformer for a Case Study in Image Fusion

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Fu, Yu Xu, Tianyang Wu, Xiao-Jun Kittler, Josef Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence School of Artificial Intelligence and Computer Science Jiangnan University Wuxi 214122 China Centre for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

The Transformer architecture has witnessed a rapid development in recent years, outperforming the CNN architectures in many computer vision tasks, as exemplified by the Vision Transformers (ViT) for image classification. However, existing visual transformer models aim to extract semantic information for high-level tasks, such as classification and detection. These methods ignore the importance of the spatial resolution of the input image, thus sacrificing the local correlation information of neighboring pixels. In this paper, we propose a Patch Pyramid Transformer(PPT) to effectively address the above issues. Specifically, we first design a Patch Transformer to transform the image into a sequence of patches, where transformer encoding is performed for each patch to extract local representations. In addition, we construct a Pyramid Transformer to effectively extract the non-local information from the entire image. After obtaining a set of multi-scale, multi-dimensional, and multi-angle features of the original image, we design the image reconstruction network to ensure that the features can be reconstructed into the original input. To validate the effectiveness, we apply the proposed Patch Pyramid Transformer to image fusion tasks. The experimental results demonstrate its superior performance, compared to the state-of-the-art fusion approaches, achieving the best results on several evaluation indicators. Thanks to the underlying representational capacity of the PPT network, it can directly be applied to different image fusion tasks without redesigning or retraining the network. Copyright © 2021, The Authors. All rights reserved.

关键词： Deep learning

Disentangled Generation Network for Enlarged License Plate Recognition and A Unified Dataset

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Li, Chenglong Yang, Xiaobin Wang, Guohao Zheng, Aihua Tan, Chang Jia, Ruoran Tang, Jin Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Artificial Intelligence Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Key Lab of Intelligent Computing and Signal Processing Ministry of Education Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China iFLYTEK Co. Ltd. Hefei230088 China

License plate recognition plays a critical role in many practical applications, but license plates of large vehicles are difficult to be recognized due to the factors of low resolution, contamination, low illumination, and occlusion, to name a few. To overcome the above factors, the transportation management department generally introduces the enlarged license plate behind the rear of a vehicle. However, enlarged license plates have high diversity as they are non-standard in position, size, and style. Furthermore, the background regions contain a variety of noisy information which greatly disturbs the recognition of license plate characters. Existing works have not studied this challenging problem. In this work, we first address the enlarged license plate recognition problem and contribute a dataset containing 9342 images, which cover most of the challenges of real scenes. However, the created data are still insufficient to train deep methods of enlarged license plate recognition, and building large-scale training data is very time-consuming and high labor cost. To handle this problem, we propose a novel task-level disentanglement generation framework based on the Disentangled Generation Network (DGNet), which disentangles the generation into the text generation and background generation in an end-to-end manner to effectively ensure diversity and integrity, for robust enlarged license plate recognition. Extensive experiments on the created dataset are conducted, and we demonstrate the effectiveness of the proposed approach in three representative text recognition frameworks. © 2022, CC BY.

关键词： Optical character recognition

Chinese Frame Disambiguation Method Based on GCN and Gate Mechanism 21

学校读者我要写书评

暂无评论

Chinese Frame Disambiguation Method Based on GCN and Gate Me...

21st Chinese National Conference on computational Linguistic, CCL 2022

作者： You, Yanan Li, Ru Su, Xuefeng Yan, Zhichao Sun, Minshuai Wang, Chao School of Computer and Information Technology Shanxi University Shanxi Taiyuan030006 China Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education Shanxi University Shanxi Taiyuan030006 China School of Modern Logistics Shanxi Vocational University of Engineering Science and Technology Shanxi Jinzhong030609 China

Chinese frame disambiguation aims to select a frame that matches its semantic scene for the target word in the sentence among the candidate *** current research methods have the defects that the calculation of the hidden layer vector has nothing to do with the target word, and ignores the influence of the syntactic structure information on the frame disambiguation. Aiming at the above problems, GCN is used to model the syntactic structure information;a gate mechanism is introduced to filter the noise information irrelevant to the target word in the hidden layer vector;and on this basis, a constraint mechanism is proposed to constrain the learning of the model and improve the representation vector. The model outperforms the current state-of-the-art models on the CFN, FN1.5 and FN1.7 datasets, proving the effectiveness of the method. © 2022 China National Conference on computational Linguistics Published under Creative Commons Attribution 4.0 International License.

关键词： Semantics

Question Generation Model Based on Iterative Message Passing and Sliding Windows Hierarchical Attention 20

学校读者我要写书评

暂无评论

Question Generation Model Based on Iterative Message Passing...

20th Chinese National Conference on computational Linguistics, CCL 2021

作者： Chen, Qian Gao, Xiaoying Wang, Suge Guo, Xin School of Computer and Information Technology Shanxi University Shanxi Taiyuan030006 China Key Laboratory Computational Intelligence Chinese Information Processing of Ministry of Education Shanxi University Shanxi Taiyuan030006 China

Knowledge graph question generation task is to generate related questions from a given knowledge *** recent years, the knowledge graph problem generation models mainly use RNN or Transformer to encode the knowledge graph subgraph, but those methods lose the explicit graph structure information, and the importance of local information to the node is ignored in the decoder. We proposes an Iterative Message Passing graph encoder to encode sub-graphs and obtain the explicit structured information of the subgraphs. In addition, we also use the Sliding Windows Hierarchical Attention mechanism to improve the RNN decoder and increase the importance of the local information of the sub-graphs to nodes. We conduct experiments on WQ and PQ data sets, and the performance of the model is significantly higher than that of the baseline model. © 2021 China National Conference on computational Linguistics Published under Creative Commons Attribution 4.0 International License

关键词： Knowledge graph

Document-Level Event Extraction Based on Frame Semantic Mapping and Type Awareness 21

学校读者我要写书评

暂无评论

Document-Level Event Extraction Based on Frame Semantic Mapp...

21st Chinese National Conference on computational Linguistic, CCL 2022

作者： Lu, Jiang Li, Ru Su, Xuefeng Yan, Zhichao Chen, Jiaxing School of Computer and Information Technology Shanxi University Shanxi Taiyuan030006 China Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education Shanxi University Shanxi Taiyuan030006 China School of Modern Logistics Shanxi Vocational University of Engineering Science and Technology Shanxi Jinzhong030609 China

Document-Level event extraction is the identification of its event type and event arguments from a given text. At present, the problems of sparse data and multi-event argument coupling are most in Document-Level events. Based on this, We map Chinese FrameNet with Chinese text events, and the sliding window mechanism and trigger word Paraphrase are introduced to improve the data sparseness problem of event detection. The use of a multi-event separation strategy based on type-aware labels alleviates the problem of meta-coupling. In order to improve the robustness of the model, adversarial training is further introduced. The experimental results of our approach on the DuEE-Fin and CCKS2021 datasets are significantly better than the existing methods. © 2022 China National Conference on computational Linguistics Published under Creative Commons Attribution 4.0 International License.

关键词： Extraction

Research on Element Extraction of Personified Sentences Based on Enhanced Characters 20

学校读者我要写书评

暂无评论

Research on Element Extraction of Personified Sentences Base...

20th Chinese National Conference on computational Linguistics, CCL 2021

作者： Li, Jing Wang, Suge Chen, Xin Wang, Dian School of Computer and Information Technology Shanxi University Shanxi Taiyuan030006 China Key Laboratory Computational Intelligence Chinese Information Processing of Ministry of Education Shanxi University Shanxi Taiyuan030006 China

In the appreciation questions of prose reading comprehension, the appreciation examination of anthropomorphic sentences is quite frequent. At present, the existing work only identification and extracts the ontological elements in the personification sentence, which has the problem of incomplete element extraction. Especially, when there are multiple ontologies in the sentence, the corresponding relationship between the personification word and each ontological element needs to be identified. In order to solve these problems, this paper proposes a method of identification and element extraction of personification sentences based on character feature enhancement. Using the characteristics of specific areas, the vector enhanced is represented for the sentence, the conditional random field model is used to identify the ontological and anthropomorphic elements in personification sentences. On this basis, the self-attention mechanism is used to detect the relationship between the elements, information exchange of between the element synchronization and relation synchronization mechanism is used to update input of element identification and relation detection. The experimental results show that the proposed method is better than other models for the extraction of on a self-built anthropomorphic dataset. © 2021 China National Conference on computational Linguistics Published under Creative Commons Attribution 4.0 International License

关键词： Synchronization