检索结果-内蒙古大学图书馆

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Wang, Chen Liu, Yuchen Chen, Boxing Zhang, Jiajun Luo, Wei Huang, Zhongqiang Zong, Chengqing National Laboratory of Pattern Recognition Institute of Automation CAS Beijing China Machine Intelligence Technology Lab Alibaba DAMO Academy China

End-to-end Speech Translation (ST) aims at translating the source language speech into target language text without generating the intermediate transcriptions. However, the training of end-to-end methods relies on parallel ST data, which are difficult and expensive to obtain. Fortunately, the supervised data for automatic speech recognition (ASR) and machine translation (MT) are usually more accessible, making zero-shot speech translation a potential direction. Existing zero-shot methods fail to align the two modalities of speech and text into a shared semantic space, resulting in much worse performance compared to the supervised ST methods. In order to enable zero-shot ST, we propose a novel Discrete Cross-Modal Alignment (DCMA) method that employs a shared discrete vocabulary space to accommodate and match both modalities of speech and text. Specifically, we introduce a vector quantization module to discretize the continuous representations of speech and text into a finite set of virtual tokens, and use ASR data to map corresponding speech and text to the same virtual token in a shared codebook. This way, source language speech can be embedded in the same semantic space as the source language text, which can be then transformed into target language text with an MT module. Experiments on multiple language pairs demonstrate that our zero-shot ST method significantly improves the SOTA, and even performs on par with the strong supervised ST baselines. © 2022 Association for Computational Linguistics.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

CMQA: A Dataset of Conditional Question Answering with Multiple-Span Answers 29

CMQA: A Dataset of Conditional Question Answering with Multi...

引用

29th International Conference on Computational Linguistics, COLING 2022

作者： Ju, Yiming Wang, Weikang Zhang, Yuanzhe Zheng, Suncong Liu, Kang Zhao, Jun National Laboratory of Pattern Recognition Institute of Automation CAS Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Machine Learning Platform Department Tencent China

Forcing the answer of the Question Answering (QA) task to be a single text span might be restrictive since the answer can be multiple spans in the context. Moreover, we found that multi-span answers often appear with two characteristics when building the QA system for a real-world application. First, multi-span answers might be caused by users lacking domain knowledge and asking ambiguous questions, which makes the question need to be answered with conditions. Second, there might be hierarchical relations among multiple answer spans. Some recent span-extraction QA datasets include multi-span samples, but they only contain unconditional and parallel answers, which cannot be used to tackle this problem. To bridge the gap, we propose a new task: conditional question answering with hierarchical multi-span answers, where both the hierarchical relations and the conditions need to be extracted. Correspondingly, we introduce CMQA, a Conditional Multiple-span Chinese Question Answering dataset to study the new proposed task. The final release of CMQA consists of 7,861 QA pairs and 113,089 labels, where all samples contain multi-span answers, 50.4% of samples are conditional, and 56.6% of samples are hierarchical. CMQA can serve as a benchmark to study the new proposed task and help study building QA systems for real-world applications. The low performance of models drawn from related literature shows that the new proposed task is challenging for the community to solve. CMQA can be accessed at https://***/juyiming/CMQA. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Domain Knowledge

来源：评论

学校读者我要写书评

暂无评论

Advancing biological super-resolution microscopy through deep learning: a brief review

引用

Biophysics Reports 2021年第4期7卷 253-266页

作者： Tianjie Yang Yaoru Luo Wei Ji Ge Yang Institute of Biophysics Chinese Academy of Sciences College of Life Sciences University of Chinese Academy of Sciences Laboratory of Computational Biology and Machine Intelligence School of Artificial IntelligenceUniversity of Chinese Academy of Sciences National Laboratory of Pattern Recognition Institute of AutomationChinese Academy of Sciences

Biological super-resolution microscopy is a new generation of imaging techniques that overcome the ～200 nm diffraction limit of conventional light microscopy in spatial resolution. By providing novel spatial or spatiotemporal information on biological processes at nanometer resolution with molecular specificity, it plays an increasingly important role in biomedical sciences. However, its technical constraints also require trade-offs to balance its spatial resolution, temporal resolution, and light exposure of samples. Recently, deep learning has achieved breakthrough performance in many image processing and computer vision tasks. It has also shown great promise in pushing the performance envelope of biological super-resolution microscopy. In this brief review, we survey recent advances in using deep learning to enhance the performance of biological super-resolution microscopy, focusing primarily on computational reconstruction of super-resolution images. Related key technical challenges are discussed. Despite the challenges, deep learning is expected to play an important role in the development of biological super-resolution microscopy. We conclude with an outlook into the future of this new research area.

关键词： Super-resolution microscopy Image super-resolution Deep learning Image reconstruction Fluorescence microscopy

来源：评论

学校读者我要写书评

暂无评论

CCLSL: Combination of Contrastive Learning and Supervised Learning for Handwritten Mathematical Expression recognition 16th

CCLSL: Combination of Contrastive Learning and Supervised ...

引用

16th Asian Conference on Computer Vision, ACCV 2022

作者： Lin, Qiqiang Huang, Xiaonan Bi, Ning Suen, Ching Y. Tan, Jun School of Mathematics and Computational Science Sun Yat-Sen University Guangzhou510275 China Guangdong Province Key Laboratory of Computational Science Sun Yat-Sen University Guangzhou510275 China Centre for Pattern Recognition and Machine Intelligence Concordia University MontrealQCH3G 1M8 Canada

ISBN: (纸本)9783031262838

Handwritten Mathematical Expressions differ considerably from ordinary linear handwritten texts, due to their two-dimentional structures plus many special symbols and characters. Hence, HMER(Handwritten Mathematical Expression recognition) is a lot more challenging compared with normal handwriting recognition. At present, the mainstream offline recognition systems are generally built on deep learning methods, but these methods can hardly cope with HEMR due to the lack of training data. In this paper, we propose an encoder-decoder method combining contrastive learning and supervised learning(CCLSL), whose encoder is trained to learn semantic-invariant features between printed and handwritten characters effectively. CCLSL improves the robustness of the model in handwritten styles. Extensive experiments on CROHME benchmark show that without data enhancement, our model achieves an expression accuracy of 58.07% on CROHME2014, 55.88% on CROHME2016 and 59.63% on CROHME2019, which is much better than all previous state-of-the-art methods. Furthermore, our ensemble model added a boost of 2.5% to 3.4% to the accuracy, achieving the state-of-the-art performance on public CROHME datasets for the first time. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Discrete Cross-Modal Alignment Enables Zero-Shot Speech Translation

arXiv

引用

arXiv 2022年

End-to-end Speech Translation (ST) aims at translating the source language speech into target language text without generating the intermediate transcriptions. However, the training of end-to-end methods relies on parallel ST data, which are difficult and expensive to obtain. Fortunately, the supervised data for automatic speech recognition (ASR) and machine translation (MT) are usually more accessible, making zero-shot speech translation a potential direction. Existing zero-shot methods fail to align the two modalities of speech and text into a shared semantic space, resulting in much worse performance compared to the supervised ST methods. In order to enable zero-shot ST, we propose a novel Discrete Cross-Modal Alignment (DCMA) method that employs a shared discrete vocabulary space to accommodate and match both modalities of speech and text. Specifically, we introduce a vector quantization module to discretize the continuous representations of speech and text into a finite set of virtual tokens, and use ASR data to map corresponding speech and text to the same virtual token in a shared codebook. This way, source language speech can be embedded in the same semantic space as the source language text, which can be then transformed into target language text with an MT module. Experiments on multiple language pairs demonstrate that our zero-shot ST method significantly improves the SOTA, and even performs on par with the strong supervised ST baselines1 Copyright © 2022, The Authors. All rights reserved.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

An Encoder-Decoder Approach to Offline Handwritten Mathematical Expression recognition with Residual Attention 3rd

An Encoder-Decoder Approach to Offline Handwritten Mathemati...

引用

3rd International Conference on pattern recognition and Artificial intelligence, ICPRAI 2022

作者： Lin, Qiqiang Wang, Chunyi Bi, Ning Suen, Ching Y. Tan, Jun School of Mathematics Sun Yat-Sen University Guangzhou510275 China Guangdong Province Key Laboratory of Computational Science Sun Yat-Sen University Guangzhou510275 China Centre for Pattern Recognition and Machine Intelligence Concordia University MontrealQCH3G 1M8 Canada

ISBN: (纸本)9783031090363

Unlike handwritten numeral recognition and handwritten text recognition, the recognition of handwritten mathematical expressions is more difficult because of their complex two-dimensional spatial structure. Since the "watch, attend and parse (WAP)" method was proposed in 2017, encoder-decoder models have made significant progress on handwritten mathematical expression recognition. Our model is improved based on the WAP [1] model. In this paper, the attention module is reasonably added to the encoder to make the extracted features are more informative. The new network is called Dense Attention Network (DATNet), which allows for an adequate extraction of the structural information from handwritten mathematical expressions. To prevent the model from overfitting during the training process, we use label smoothing [2] as the loss. Experiments showed that our model (DATWAP) improved WAP expression recognition from 48.4%/46.8%/45.1% to 54.72%/52.83%/52.54% on the CROHME 2014/2016/2019 test sets. © 2022, Springer Nature Switzerland AG.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

SIR-HCL: Semantic-Inconsistency Reasoning and Hybrid Contrastive Learning for Efficient Cross-Emotion Anomaly Detection

引用

IEEE Transactions on Cognitive and Developmental Systems 2025年

作者： Liu, Xin Chen, Qiyan Cheung, Yiu-Ming Peng, Shu-Juan Huaqiao University Department of Computer Science Xiamen361021 China Hong Kong Baptist University Department of Computer Science SAR Hong Kong Hong Kong Xiamen Key Laboratory of Computer Vision and Pattern Recognition Xiamen361021 China Huaqiao University Fujian Key Laboratory of Big Data Intelligence and Security Xiamen361021 China Huaqiao University Department of Artificial Intelligence Xiamen China Fujian Province University Key Laboratory of Computer Vision and Machine Learning Huaqiao University Xiamen361021 China

Cross-emotion anomaly detection is an emerging and challenging research topic in cognitive analysis field, which aims at identifying the abnormal emotion pair whose semantic patterns are inconsistent across different emotional modalities. To the best of our knowledge, this topic has yet to be well studied, which could potentially benefit lots of valuable cognitive applications such as autistic children diagnosis and criminal deception detection. To this end, this paper proposes an efficient cross-emotion anomaly detection approach via semanticinconsistency reasoning and hybrid contrastive learning (SIR-HCL), which is the first attempt to detect the anomalous emotional pairs across the audio-visual emotions. First, the proposed framework utilizes dual-branch network to obtain the deep emotional features in each modality, and then employs the shared residual block to derive the semantically compatible features. Subsequently, an efficient hybrid contrastive learning approach is designed to enlarge the semantic-inconsistency among abnormal emotional pair with different affective classes, while enhancing the semantic-consistency and increasing the feature correlation between normal emotional pair from the same affective class. At the same time, an efficient bidirectional learning scheme is employed to significantly improve the data utilization and a two-component Beta Mixture Model is adaptively utilized to reason the anomalous emotion pairs. Extensive experiments evaluated on two benchmark datasets show that the proposed SIR-HCL method can well detect the anomalous emotional pairs across audio-visual emotional data, and brings substantial improvements over the state-of-the-art competing methods. © 2016 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning in Palmprint recognition-A Comprehensive Survey

arXiv

引用

arXiv 2025年

作者： Gao, Chengrui Yang, Ziyuan Jia, Wei Leng, Lu Zhang, Bob Teoh, Andrew Beng Jin College of Computer Science Sichuan University Chengdu610065 China Singapore School of Computer and Information Hefei University of Technology Hefei China Jiangxi Provincial Key Laboratory of Image Processing and Pattern Recognition Nanchang Hangkong University Nanchang China Pattern Analysis and Machine Intelligence Group Department of Computer and Information Science University of Macau Taipa China School of Electrical and Electronic Engineering College of Engineering Yonsei University Seoul Korea Republic of

Palmprint recognition has emerged as a prominent biometric technology, widely applied in diverse scenarios. Traditional handcrafted methods for palmprint recognition often fall short in representation capability, as they heavily depend on researchers' prior knowledge. Deep learning (DL) has been introduced to address this limitation, leveraging its remarkable successes across various domains. While existing surveys focus narrowly on specific tasks within palmprint recognition-often grounded in traditional methodologies-there remains a significant gap in comprehensive research exploring DL-based approaches across all facets of palmprint recognition. This paper bridges that gap by thoroughly reviewing recent advancements in DL-powered palmprint recognition. The paper systematically examines progress across key tasks, including region-of-interest segmentation, feature extraction, and security/privacy-oriented challenges. Beyond highlighting these advancements, the paper identifies current challenges and uncovers promising opportunities for future research. By consolidating state-of-the-art progress, this review serves as a valuable resource for researchers, enabling them to stay abreast of cutting-edge technologies and drive innovation in palmprint recognition. Copyright © 2025, The Authors. All rights reserved.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Offline Handwritten Mathematical Expression recognition Based on Square Loss Sequence-to-Sequence Neural Network

SSRN

引用

SSRN 2022年

作者： Wang, Fanglin He, Fukeng Bi, Ning Suen, Ching Y. Tan, Jun School of Mathematics Sun Yat-sen University China Tencent China Guangdong Province Key Laboratory of Computational Science Sun Yat-Sen University China Centre for Pattern Recognition and Machine Intelligence Concordia University Canada

Since mathematical expressions play an indispensable role in scientific literature, handwritten mathematical expression recognition (HMER) has recently become an important topic that helps scientists effortlessly use mathematical formulas in their theses and other documents. However, the accuracy of offline handwritten mathematical recognition is currently limited by the ambiguity of mathematical symbols, the complex two-dimensional structure of mathematical formulas and the strong dependence on context. In this paper, we propose a novel offline handwritten mathematical recognition approach based on square loss encoderdecoder neural network (SLSSN). In the proposed neural network, a square loss module, a pre-training module and a statistical language model are added, which can effectively improve the accuracy of offline HMER. Our method’s single model recognition rate exceeds the current best model on the CROHME 2014 test set, and the overall recognition rates for formulas with less than 1, 2 and 3 errors reach high levels of 68.25 %, 76.37 % and 83.16 %, respectively. © 2022, The Authors. All rights reserved.

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

Cclsl: Combination of Contrastive Learning and Supervised Learning for Handwritten Mathematical Expression recognition

SSRN

引用

SSRN 2023年

作者： Lin, Qiqiang Xiao, Haiyang Bi, Ning Suen, Ching Y. Tan, Jun School of Mathematics Sun Yat-Sen University Guangzhou510275 China Guangdong Province Key Laboratory of Computational Science Sun Yat-Sen University Guangzhou510275 China Centre for Pattern Recognition and Machine Intelligence Concordia University MontrealQCH3G 1M8 Canada

Handwritten Mathematical Expressions differ considerably from ordinary linear handwritten texts, due to their two-dimentional structures plus many special symbols and characters. Hence, HMER(Handwritten Mathematical Expression recognition) is a lot more challenging compared with normal handwriting recognition. At present, the mainstream offline recognition systems are generally built on deep learning methods, but these methods can hardly cope with HEMR due to the lack of training data and diverse writing styles. In this paper, we propose a pre-training framework combining contrastive learning and supervised learning(CCLSL), whose encoder is trained to learn semantic-invariant features between printed and handwritten characters effectively. CCLSL improves the robustness of the model in handwritten styles. Extensive experiments on CROHME benchmark show that without data enhancement, our model achieves an expression accuracy of 58.07% on CROHME2014, 55.88% on CROHME2016 and 59.63% on *** fine-tuning on the handwritten dataset, the accuracy of the model on CROHME2014, CROHME2016, and CROHME2019 reached 59.89%, 56.49%, and 60.13%, which is much better than all previous state-of-the-art methods. © 2023, The Authors. All rights reserved.

关键词： Semantics

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：