检索结果-内蒙古大学图书馆

Rhythm-Aware sequence-to-sequence Learning for Labanotation Generation With Gesture-Sensitive Graph Convolutional Encoding

引用

IEEE TRANSACTIONS ON MULTIMEDIA 2022年 24卷 1488-1502页

作者： Li, Min Miao, Zhenjiang Zhang, Xiao-Ping Xu, Wanru Ma, Cong Xie, Ningwei Beijing Jiaotong Univ Inst Informat Sci Beijing 100044 Peoples R China Ryerson Univ Dept Elect Comp & Biomed Engn Toronto ON M5B 2K3 Canada

Labanotation is a professional dance notation system widely used in dance education and choreography preservation. Automatically generatingLabanotation dance scores from motion capture data can save a huge amount of manual time and effort. Recently, the sequence-to-sequence (seq2seq) model is applied to the automatic Labanotation generation. This model is based on an encoder-decoder structure, which encodes the input motion sequence to a fixed-length vector and then decodes it to generate the target sequence. However, the encoding of spatial skeleton structure of motion data is not considered in the existing work. Besides, it is challenging to align between the input motion data and the output Laban symbol sequences due to the severe imbalance of sequence lengths. Therefore, in this paper, we present a new seq2seq model for more effective Labanotation generation. In the encoder, we propose a new gesture-sensitive graph convolutional network with learned adaptive joint weights and non-physical connections to learn both spatial and temporal patterns from motion data sequences. In the decoder, we exploit motion rhythm information and propose a novel rhythm-aware attention mechanism to learn a good alignment between motion sequences and Laban symbol sequences, so that we can focus on relevant parts of the input motion sequence without searching in the whole sequence when predicting a target Laban symbol. Extensive experiments on two real-world datasets show that the proposed method achieves a better performance compared with the state-of-the-art approaches on the task of automatic Labanotation generation.

关键词： Motion segmentation Data models Skeleton Hidden Markov models Feature extraction Decoding Rhythm Attention mechanism graph convolutional network labanotation generation sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

Improving Rumor Detection by Image Captioning and Multi-Cell Bi-RNN With Self-Attention in Social Networks

引用

INTERNATIONAL JOURNAL OF DATA WAREHOUSING AND MINING 2022年第1期18卷 1页

作者： Wang, Jenq-Haur Huang, Chin-Wei Norouzi, Mehdi Natl Taipei Univ Technol Taipei Taiwan Univ Cincinnati Cincinnati OH USA

User-generated contents in social media are not verified before being posted. They could bring many problems if they were misused. Among various types of rumors, the authors focus on the type in which there's mismatch between images and their surrounding texts. They can be detected by multimodal feature fusion in RNNs with attention mechanism, but the relations between images and texts are not well-addressed. In this paper, the authors propose to improve rumor detection by image captioning and RNNs with self-attention. Firstly, they utilize the idea of image captioning to translate images into the corresponding text descriptions. Secondly, these caption words are represented by word embedding models and aggregated with surrounding texts using early fusion. Finally, multi-cell bidirectional RNNs with self-attention are used to learn important features to identify rumors. From the experimental results, the best F-measure of 0.882 can be obtained, which shows the potential of our proposed approach to rumor detection. Further investigation is needed for data in larger scale.

关键词： Bi-Directional Gated Recurrent Unit Multimodal Feature Fusion Rumor Detection Image Captioning Self-Attention Mechanism sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

A Non-Autoregressive Neural Machine Translation model With Iterative Length Update of Target Sentence

引用

IEEE ACCESS 2022年 10卷 43341-43350页

作者： Lim, Yeon-Soo Park, Eun-Ju Song, Hyun-Je Park, Seong-Bae Kyung Hee Univ Dept Comp Sci & Engn Yonin Si 17104 Gyeonggi Do South Korea Jeonbuk Natl Univ Dept Informat & Engn Jeonju Si 54896 Jeollabuk Do South Korea

The non-autoregressive decoders in neural machine translation are paid increasing attention due to their faster decoding than autoregressive decoders. However, their apparent problem is a low performance which is mainly originated from wrong prediction about the target sentence length. To attack this problem, this paper proposes a novel machine translation model with a new non-autoregressive decoder named Iterative and Length-Adjustive Non-Autoregressive Decoder (ILAND). This decoder adopts a masked language model to avoid generation of low-confident tokens and changes the length of a target sentence iteratively to an optimal length. To complete these goals, ILAND consists of three complementary sub-modules of a token masker, a length adjuster, and a token generator. The token masker and the token generator take charge of the masked language model, and the length adjuster optimizes the target sentence length. The sequence-to-sequence training of the translation model is also introduced. In this training, the length adjuster and the token generator are jointly trained since they share a similar structure. The effectiveness of the translation model is proven by showing empirically that the model outperforms other models with various non-autoregressive decoders. The thorough analysis suggests that the performance gain of the translation model comes from target sentence length adaptation and the joint learning. In addition, ILAND is also shown to be faster than other iterative non-autoregressive decoders while it is still robust against the multi-modality problem.

关键词： Decoding Iterative decoding Generators Adaptation models Machine translation Predictive models Transformers Machine translation non-autoregressive decoder sequence-to-sequence model target length adaptation transformer

来源：评论

学校读者我要写书评

暂无评论

TBCUP: A Transformer-based Code Comments Updating Approach 47

TBCUP: A Transformer-based Code Comments Updating Approach

引用

47th IEEE-Computer-Society Annual International Conference on Computers, Software, and Applications (COMPSAC)

作者： Liu, Shifan Cui, Zhanqi Chen, Xiang Yang, Jun Li, Li Zheng, Liwei Beijing Informat Sci & Technol Univ Sch Comp Sci Beijing Peoples R China Nantong Univ Sch Informat Sci & Technol Nantong Peoples R China

ISBN: (纸本)9798350326970

Good code comments are of great value for program maintenance. However, during the development process, developers often neglect to update the corresponding comments when changing the code, which results in inconsistent comments and affects the maintainability of software. Studies have shown that even in widely used programs, there is a large number of outdated comments. Existing code comment update approaches use long short-term memory (LSTM) model based on encoder-decoder to capture the relationship between code changes and comment updates, to automatically update code comments. However, due to the complexity of code changes, the corresponding comments update results do not perform well. Thus, a Transformer-based automatic update approach for code comments called TBCUP is proposed in this paper. TBCUP uses the multi-head attention mechanism to learn the relationship between code change sequences and comment updates to update comments more accurately. In addition, TBCUP uses the Byte Pair Encoding (BPE) algorithm to build a unified vocabulary for code and corresponding comments to alleviate the problem of out-of-vocabulary (OOV) words while updating comments. With BPE, TBCUP can split or combine words to handle OOV words. The experimental results on the dataset containing 108k sets of code-comment co-evolution samples show that the accuracy of TBCUP is improved by 3.2% in comparison with CUP.

关键词： program comprehension comment updating deep learning sequence-to-sequence model code changes

来源：评论

学校读者我要写书评

暂无评论

Abstractive Text Summarization Using Multimodal Information 10

Abstractive Text Summarization Using Multimodal Information

引用

10th International Conference on Soft Computing and Machine Intelligence (ISCMI)

作者： Rafi, Shaik Das, Ranjita NIT MIZORAM Dept Comp Sci & Engn Aizawl 796012 Mizoram India NIT AGARTALA Dept Comp Sci & Engn Agartala 799046 Tripura India

ISBN: (纸本)9798350359374

Much text generates over the internet through news articles, story writing and blogs. Reading and understanding such an enormous amount of data to the user is problematic, including time and effort. Automatic abstractive text summarization has gained more importance to increase the user's understanding and reduce time. It shortens the given input by preserving the meaning and identifying the context of the whole document to generate meaningful sentences. The research community has proposed different methods for text reduction and generating abstractive summaries. However, problems like semantics and contextual relationship in the summary generation process must be still need to improve. The multimodal abstractive text summarization is a technique that combines text and image information which helps in addressing the semantics and contextual relationship by proposing Multimodality Image Text (MIT) layer that fuses the text-extracted global features by glove embedding and preserves the semantics of the vocabulary and text-related images are used to identify the contextual relationship features from inception v3, which cope in the MIT layer to generate efficient multimodal abstractive text summaries by training and testing with seq-to-seq model. Experiments with the MSMO dataset achieve superior performance on other state-of-art results.

关键词： Abstractive Text Summarization Multimodality Image Text (MIT) Attention Mechanism LSTM sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

Singing Voice Synthesis Based on a Musical Note Position-Aware Attention Mechanism 48

Singing Voice Synthesis Based on a Musical Note Position-Awa...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Hono, Yukiya Hashimoto, Kei Nankaku, Yoshihiko Tokuda, Keiichi Nagoya Institute of Technology Nagoya Japan

ISBN: (纸本)9781728163277

This paper proposes a novel sequence-to-sequence (seq2seq) model with a musical note position-aware attention mechanism for singing voice synthesis (SVS). A seq2seq modeling approach that can simultaneously perform acoustic and temporal modeling is attractive. However, due to the difficulty of the temporal modeling of singing voices, many recent SVS systems with an encoder-decoder-based model still rely on explicitly on duration information generated by additional modules. Although some studies perform simultaneous modeling using seq2seq models with an attention mechanism, they have insufficient robustness against temporal modeling. The proposed attention mechanism is designed to estimate the attention weights by considering the rhythm given by the musical score. Furthermore, several techniques are also introduced to improve the modeling performance of the singing voice. Experimental results indicated that the proposed model is effective in terms of both naturalness and robustness of timing. © 2023 IEEE.

关键词： attention mechanism sequence-to-sequence model Singing voice synthesis temporal modeling

来源：评论

学校读者我要写书评

暂无评论

Combining Transformers and Tree-based Decoders for Solving Math Word Problems

Combining Transformers and Tree-based Decoders for Solving M...

引用

2023 IEEE International Conference on Big Data, BigData 2023

作者： Wang, Jenq-Haur Chen, Yi-Fang National Taipei University of Technology Department of Computer Science and Information Engineering Taipei Taiwan Engineer *** Inc Taipei Taiwan

ISBN: (纸本)9798350324457

Solving math word problems is a popular topic in natural language processing. We not only need to classify the grammatical structures in the questions, but also understand the mathematical logic expressed between words. Errors in semantic understanding may lead to the failure to generate correct solution equations. Thus, the correct answer cannot be calculated. Previous studies mainly used sequence-to-sequence recurrent neural networks (RNNs) to obtain meaning in words, or combined graph neural networks to capture more information in questions to achieve better results. In addition, recent studies also showed that, a tree-based decoder leads to better results than a decoder of RNNs. In this paper, we propose to combine transformers and tree-based decoders for solving math word problems. Firstly, we use a transformer encoder to read math word problems, whose outputs are given to two different decoders, including Transformer decoder, and a treebased decoder. Secondly, from the answer equations generated from the two decoders, the better solution is selected. The experimental results on the two commonly used math word problem datasets, MAWPS and ASDiv-A, show that our model achieves 89.3% and 81.9% accuracy, which are 2.2% and 4.2% higher than the vanilla transformer model, respectively. For the MAWPS dataset the performance is comparable to state-of-the-art model Graph2Tree. This shows the effectiveness of our proposed method. © 2023 IEEE.

关键词： Math Word Problem Self-Attention Mechanism sequence-to-sequence model Transformer Tree-based Decoder

来源：评论

学校读者我要写书评

暂无评论

sequence-to-sequence Knowledge Graph Completion Based on Gated Attention Unit 29

Sequence-to-Sequence Knowledge Graph Completion Based on Gat...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Yi, Fengge Wei, Xiumei Liu, Xiaojing Jiang, Xuesong Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan China State Key Laboratory of High-end Server & Storage Technology Jinan China

ISBN: (纸本)9798350330717

We present GauKGT5, a sequence-to-sequence model proposed for knowledge graph completion (KGC). Our research extends the KGT5 model, a recent sequence-to-sequence link prediction (LP) model. GauKGT5 takes advantage of textual characteristics inherent in the knowledge graph, exhibiting a small model size. However, KGT5's proficiency in link prediction necessitates the ensemble with a knowledge graph embedding model, which itself poses challenges due to its substantial size and expense. By integrating the Gated Attention Unit into the KGT5 model and directly applying it to the encoder-decoder structure, we achieve improved contextual dependency capturing within the sequence, resulting in enhanced prediction accuracy, accelerated training speed, and enhanced computational efficiency. At the same time, we introduce parallel computing as a means to enhance the efficiency of model training and inference within the XPU distributed computing environment. © 2023 IEEE.

关键词： gated attention unit knowledge graph completion sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

Leveraging Causal Inference for Explainable Automatic Program Repair

Leveraging Causal Inference for Explainable Automatic Progra...

引用

IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) / IEEE World Congress on Computational Intelligence (IEEE WCCI) / International Joint Conference on Neural Networks (IJCNN) / IEEE Congress on Evolutionary Computation (IEEE CEC)

作者： Wang, Jianzong Si, Shijing Zhu, Zhitao Qu, Xiaoyang Hong, Zhenhou Xiao, Jing Ping An Technol Shenzhen Co Ltd Shenzhen Peoples R China

ISBN: (数字)9781728186719

ISBN: (纸本)9781728186719

Deep learning models have made significant progress in automatic program repair. However, the black-box nature of these methods has restricted their practical applications. To address this challenge, this paper presents an interpretable approach for program repair based on sequence-to-sequence models with causal inference and our method is called CPR, short for causal program repair. Our CPR can generate explanations in the process of decision making, which consists of groups of causally related input-output tokens. Firstly, our method infers these relations by querying the model with inputs disturbed by data augmentation. Secondly, it generates a graph over tokens from the responses and solves a partitioning problem to select the most relevant components. The experiments on four programming languages (Java, C, Python, and JavaScript) show that CPR can generate causal graphs for reasonable interpretations and boost the performance of bug fixing in automatic program repair.

关键词： Automated Program Repair Program Analysis sequence-to-sequence model Causal Inference Interpretability

来源：评论

学校读者我要写书评

暂无评论

A Conformer-basedWaveform-domain Neural Acoustic Echo Canceller Optimized for ASR Accuracy 23

A Conformer-basedWaveform-domain Neural Acoustic Echo Cancel...

引用

Interspeech Conference

作者： Panchapagesan, Sankaran Narayanan, Arun Shabestary, Turaj Zakizadeh Shao, Shuai Howard, Nathan Park, Alex Walker, James Gruenstein, Alexander Google LLC Mountain View CA 94043 USA

Acoustic Echo Cancellation (AEC) is essential for accurate recognition of queries spoken to a smart speaker that is playing out audio. Previous work has shown that a neural AEC model operating on log-mel spectral features (denoted "logmel" hereafter) can greatly improve Automatic Speech Recognition (ASR) accuracy when optimized with an auxiliary loss utilizing a pre-trained ASR model encoder. In this paper, we develop a conformer-based waveform-domain neural AEC model inspired by the "TasNet" architecture. The model is trained by jointly optimizing Negative Scale-Invariant SNR (SISNR) and ASR losses on a large speech dataset. On a realistic rerecorded test set, we find that cascading a linear adaptive AEC and a waveform-domain neural AEC is very effective, giving 56-59% word error rate (WER) reduction over the linear AEC alone. On this test set, the 1.6M parameter waveform-domain neural AEC also improves over a larger 6.5M parameter logmel-domain neural AEC model by 20-29% in easy to moderate conditions. By operating on smaller frames, the waveform neural model is able to perform better at smaller sizes and is better suited for applications where memory is limited.

关键词： Acoustic echo cancellation TasNet Waveform Neural AEC sequence-to-sequence model multi-task loss

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：