检索结果-内蒙古大学图书馆

32nd ACM World Wide Web Conference, WWW 2023

作者： Zhang, Ke Wang, Xiaoqing Cheng, Gong State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

ISBN: (纸本)9781450394161

The Diameter-bounded max-Coverage Group Steiner Tree (DCGST) problem has recently been proposed as an expressive way of formulating keyword-based search and exploration of knowledge graphs. It aims at finding a diameter-bounded tree which covers the most given groups of vertices and has the minimum weight. In contrast to its specialization - the classic Group Steiner Tree (GST) problem which has been extensively studied, the emerging DCGST problem still lacks an efficient algorithm. In this paper, we propose Cba, the first approximation algorithm for the DCGST problem, and we prove its worst-case approximation ratio. Furthermore, we incorporate a best-first search strategy with two pruning methods into PrunedCBA, an improved approximation algorithm. Our extensive experiments on real and synthetic graphs demonstrate the effectiveness and efficiency of PrunedCBA. © 2023 ACM.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Understanding Bugs in Rust Compilers 23

Understanding Bugs in Rust Compilers

引用

23rd IEEE International Conference on software Quality, Reliability, and Security, QRS 2023

作者： Xia, Xinmeng Feng, Yang Shi, Qingkai Nanjing University State Key Laboratory for Novel Software Technology Nanjing China

ISBN: (纸本)9798350319583

Rust compilers play a foundational role in the Rust language. Like any complex system, they are susceptible to bugs, which can impact the correctness and reliability of the compiled Rust programs. To gain a deeper understanding of these bugs, this paper presents the first comprehensive analysis of historical bugs in two widely used Rust compilers: Rustc and Rust-GCC. The analysis delves into the bugs' characteristics, bug-proneness locations, bug root causes, and bug-fixing efforts. The findings reveal that the majority of bugs in Rustc are associated with the compiler's kernel, while Rust-GCC experiences most bugs related to the cleanup process. Among all modules, the 'src/librustc' module exhibits the highest bug-proneness in the Rustc compiler, whereas the 'gcc/rust' modules demonstrate the highest bug-proneness in the Rust-GCC compiler. Furthermore, the study reveals that the bug-fixing process is accelerated when test cases utilize Rust's concurrency features. © 2023 IEEE.

关键词： Program compilers

来源：评论

学校读者我要写书评

暂无评论

Speech Recognition Method Based on Deep Learning of Artificial Intelligence: An example of BLSTM-CTC model 2023

Speech Recognition Method Based on Deep Learning of Artifici...

引用

5th International Symposium on Signal Processing Systems, SSPS 2023

作者： Chen, Kangyu Peng, Zhiyuan Department of Computer Science The University of Hong Kong Hong Kong State Key Laboratory for Novel Software Technology Nanjing University China

ISBN: (纸本)9798400700040

Under the influence of information, network and intelligent high-speed development situation, China's intelligent technology and other aspects have made great progress and achievements, derived a lot of advanced artificial intelligence technology, machine learning technology and deep learning technology, etc., to promote the development of intelligence and information in major fields. Artificial intelligence deep learning is the fusion of artificial intelligence technology and machine learning technology, which lays the foundation for the reform and innovation of artificial voice intelligent recognition technology and intelligent robot technology. So in order to improve the application level of intelligent speech recognition technology, it is necessary to continuously optimize the speech recognition method based on AI deep learning. In this regard, according to the relevant literature, this paper addresses the problem that phoneme features of varying duration are generated during the propagation of speech signals, and these features affect the correct rate of speech recognition, and the phoneme features of different lengths are standardized based on the deep learning research mentioned in this paper with BLSTM-CTC as an example. By evaluating the model on the Thchs30 and ST-CMDS datasets, the results show that the MCFN-based BLSTM-CTC speech recognition model has a reduced recognition word error rate compared with the traditional speech recognition model. © 2023 ACM.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

AutoQNN: An End-to-End Framework for Automatically Quantizing Neural Networks

引用

Journal of computer science & technology 2024年第2期39卷 401-420页

作者：龚成卢冶代素蓉邓倩杜承昆李涛 College of Software Nankai UniversityTianjin 300350China College of Computer Science Nankai UniversityTianjin 300350China State Key Laboratory of Processors Institute of Computing TechnologyChinese Academy of Sciences Beijing 100190China

Exploring the expected quantizing scheme with suitable mixed-precision policy is the key to compress deep neural networks(DNNs)in high efficiency and *** exploration implies heavy workloads for domain experts,and an automatic compression method is ***,the huge search space of the automatic method introduces plenty of computing budgets that make the automatic process challenging to be applied in real *** this paper,we propose an end-to-end framework named AutoQNN,for automatically quantizing different layers utilizing different schemes and bitwidths without any human *** can seek desirable quantizing schemes and mixed-precision policies for mainstream DNN models efficiently by involving three techniques:quantizing scheme search(QSS),quantizing precision learning(QPL),and quantized architecture generation(QAG).QSS introduces five quantizing schemes and defines three new schemes as a candidate set for scheme search,and then uses the Differentiable Neural Architecture Search(DNAS)algorithm to seek the layer-or model-desired scheme from the *** is the first method to learn mixed-precision policies by reparameterizing the bitwidths of quantizing schemes,to the best of our *** optimizes both classification loss and precision loss of DNNs efficiently and obtains the relatively optimal mixed-precision model within limited model size and memory *** is designed to convert arbitrary architectures into corresponding quantized ones without manual intervention,to facilitate end-to-end neural network *** have implemented AutoQNN and integrated it into *** experiments demonstrate that AutoQNN can consistently outperform state-of-the-art *** 2-bit weight and activation of AlexNet and ResNet18,AutoQNN can achieve the accuracy results of 59.75%and 68.86%,respectively,and obtain accuracy improvements by up to 1.65%and 1.74%,respectively,compared with state-of-the-art ***,c

关键词： automatic quantization mixed precision quantizing scheme search quantizing precision learning quan-tized architecture generation

来源：评论

学校读者我要写书评

暂无评论

DyFuzz: Skeleton-based Fuzzing for Python Libraries 23

DyFuzz: Skeleton-based Fuzzing for Python Libraries

引用

23rd IEEE International Conference on software Quality, Reliability, and Security, QRS 2023

作者： Xia, Xinmeng Feng, Yang Nanjing University State Key Laboratory for Novel Software Technology Nanjing China

ISBN: (纸本)9798350319583

Programming libraries are indispensable for programming languages. Programmers can access the pre-written codes in these libraries via the application programmable interfaces (API), optimizing and accelerating their programming tasks. However, defects in these libraries may cause unexpected software behaviors, threatening their robustness and safety. Thus, it is crucial to ensure the quality of the libraries. This paper explores an alternative approach, namely Fuzzing Skeleton API (FSA), for detecting library bugs in Python. For the given API, FSA aims to generate massive inputs, i.e., different argument combinations, and pass them to the API to verify its correctness and reliability. To realize this, FSA first abstracts the API into a skeleton by modeling its usage of parameters as placeholders. Then, it can generate the seed API calls by filling these placeholders with pre-defined arguments. Finally, the approach incorporates four mutation strategies, i.e., bit mutation, literal mutation, element mutation, and attribute mutation, to mutate different arguments and hence generate massive API calls. We have implemented the proposed approach into an automated tool, namely DyFuzz, for testing Python libraries. In less than one month of the fuzzing experiment, DyFuzz detected 14 library bugs, of which nine have been confirmed as unknown bugs. © 2023 IEEE.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

DenseNet-based RFID Grouping Protocols

DenseNet-based RFID Grouping Protocols

引用

2023 International Conference on Artificial Intelligence of Things and Systems, AIoTSys 2023

作者： Wang, Tianyu Liu, Jia Nanjing University State Key Laboratory for Novel Software Technology Nanjing China

ISBN: (纸本)9798350312270

The grouping protocol in RFID systems is to label tags according to a given partition so that tags in the identical group hold the same group ID, which makes multi-cast transmissions or aggregate queries possible and thereby improves time efficiency. Existing grouping protocols need to inform each tag individually or each group by transmitting extra filter vectors, which suffer degraded performance. In this paper, we propose a DenseNet-based grouping protocol that labels tags concurrently. The basic idea is to treat the grouping problem as a classification task. The partition information known by an RFID reader is used to train data and obtain a classifier, of which parameters can be broadcast to all tags simultaneously. Each tag can calculate its own group ID according to the parameters. Experimental results demonstrate that our grouping protocol can reduce the execution time by 53% in a real-world RFID system with about 100,000 tags, compared with the state-of-the-art. © 2023 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Bi-Directional and Triangular Circulation Fusion Neural Networks for Small Object Detection

引用

IEEE Transactions on Circuits and Systems for Video technology 2024年第6期35卷 5140-5152页

作者： Li, Fangyu Duan, Junzhu Zhang, Qiyu Shan, Caifeng Han, Honggui the School of Information Science and Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Engineering Research Center of Digital Community Ministry of Education Beijing Artificial Intelligence Institute Beijing University of Technology Beijing100124 China the State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China the School of Intelligence Science and Technology Nanjing University Suzhou215163 China

Deep learning-driven object detection models are capable of accurately identifying and localizing objects. However, small objects contain limited information relative to global features, resulting in the fact that detection models often do not learn small object features adequately. To enhance the precision in detecting small objects, we propose a bi-directional and triangular circulation fusion neural network (BTFN). First, to selectively strengthen the position features of small objects, we propose a feature circulation extraction module composed of a bi-directional triangular densely nested convolutional network (BTF), thus achieving repetitive multi-layer feature fusion. Second, to fill up the semantic gaps between different scales of features, we design a mixed dual attention module (MDA) in the bi-directional triangular densely nested network. Third, to mitigate the lost information in the neural networks with deep layers as well as improve the inference time, we design a re-parameterization bi-directional composite feature fusion module (Rep-BFM) that fuses the features of multiple scales. The proposed model is evaluated extensively on the MS COCO, Tsinghua-Tencent 100k, and Haier dismantled parts of used home appliances datasets. The experiment results show that the proposed model improves the AP on MS COCO by 4%, especially the APS of small objects is improved by 7.7% compared with SOTA models. © 2024 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

关键词： Domestic appliances

来源：评论

学校读者我要写书评

暂无评论

Learning Robust Multi-Modal Representation for Multi-Label Emotion Recognition via Adversarial Masking and Perturbation 23

Learning Robust Multi-Modal Representation for Multi-Label E...

引用

32nd ACM World Wide Web Conference, WWW 2023

作者： Ge, Shiping Jiang, Zhiwei Cheng, Zifeng Wang, Cong Yin, Yafeng Gu, Qing State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

ISBN: (纸本)9781450394161

Recognizing emotions from multi-modal data is an emotion recognition task that requires strong multi-modal representation ability. The general approach to this task is to naturally train the representation model on training data without intervention. However, such natural training scheme is prone to modality bias of representation (i.e., tending to over-encode some informative modalities while neglecting other modalities) and data bias of training (i.e., tending to overfit training data). These biases may lead to instability (e.g., performing poorly when the neglected modality is dominant for recognition) and weak generalization (e.g., performing poorly when unseen data is inconsistent with overfitted data) of the model on unseen data. To address these problems, this paper presents two adversarial training strategies to learn more robust multi-modal representation for multi-label emotion recognition. Firstly, we propose an adversarial temporal masking strategy, which can enhance the encoding of other modalities by masking the most emotion-related temporal units (e.g., words for text or frames for video) of the informative modality. Secondly, we propose an adversarial parameter perturbation strategy, which can enhance the generalization of the model by adding the adversarial perturbation to the parameters of model. Both strategies boost model performance on the benchmark MMER datasets CMU-MOSEI and NEMu. Experimental results demonstrate the effectiveness of the proposed method compared with the previous state-of-the-art method. Code will be released at https://***/ShipingGe/MMER. © 2023 ACM.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

TalkingStyle: Personalized Speech-Driven 3D Facial Animation with Style Preservation

引用

IEEE Transactions on Visualization and computer Graphics 2024年 PP卷 1-12页

作者： Song, Wenfeng Wang, Xuan Zheng, Shi Li, Shuai Hao, Aimin Hou, Xia Computer School Beijing Information Science and Technology University China State Key Laboratory of Virtual Reality Technology and Systems Beihang University China

It is a challenging task to create realistic 3D avatars that accurately replicate individuals' speech and unique talking styles for speech-driven facial animation. Existing techniques have made remarkable progress but still struggle to achieve lifelike mimicry. This paper proposes “TalkingStyle”, a novel method to generate personalized talking avatars while retaining the talking style of the person. Our approach uses a set of audio and animation samples from an individual to create new facial animations that closely resemble their specific talking style, synchronized with speech. We disentangle the style codes from the motion patterns, allowing our method to associate a distinct identifier with each person. To manage each aspect effectively, we employ three separate encoders for style, speech, and motion, ensuring the preservation of the original style while maintaining consistent motion in our stylized talking avatars. Additionally, we propose a new style-conditioned transformer decoder, offering greater flexibility and control over the facial avatar styles. We comprehensively evaluate TalkingStyle through qualitative and quantitative assessments, as well as user studies demonstrating its superior realism and lip synchronization accuracy compared to current state-of-the-art methods. To promote transparency and further advancements in the field, we also make the source code publicly available at https://***/wangxuanx/TalkingStyle. IEEE

关键词： Synchronization

来源：评论

学校读者我要写书评

暂无评论

Easy Travelogue: A Travelogue Editor with Automatic Image Recommendation and Insertion 5

Easy Travelogue: A Travelogue Editor with Automatic Image Re...

引用

5th ACM International Conference on Multimedia in Asia, MMAsia 2023

作者： Yu, Fan Xing, Huanyu Bei, Jia Ren, Tongwei State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

ISBN: (纸本)9798400702051

Travelogues are a common media form that incorporates both text and images. Typically, they are composed after the completion of a travel period. Creating a travelogue demands substantial time and effort, particularly in the curation of suitable images from the extensive collection of photos taken during the journey to complement the text. Consequently, we have developed and implemented Easy Travelogue, a travelogue editor that utilizes visual and language models. It offers real-time image suggestions while writing the text and can automatically insert fitting images into the finished content. The editor is versatile and can be readily utilized for personal travelogues, travel blogs, and various social media platforms, facilitating users in effortlessly sharing and showcasing their travel experiences. © 2023 Copyright held by the owner/author(s).

关键词： Visual languages

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：