检索结果-内蒙古大学图书馆

12th National CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2023

作者： Fan, Yaxin Jiang, Feng Li, Peifeng Li, Haizhou School of Data Science The Chinese University of Hong Kong Shenzhen China School of Computer Science and Technology Soochow University Suzhou China Shenzhen Research Institute of Big Data Guangdong Shenzhen China School of Information Science and Technology University of Science and Technology of China Hefei China

ISBN: (纸本)9783031446986

Grammatical error correction aims to correct ungrammatical sentences automatically. Recently, some work has demonstrated the excellent capabilities of closed-source Large Language Models (LLMs, e.g., ChatGPT) in grammatical error correction. However, the potential of open-source LLMs remains unexplored. In this paper, we introduced GrammarGPT, an open-source LLM, to preliminary explore its potential for native Chinese grammatical error correction. The core recipe of GrammarGPT is to leverage the hybrid dataset of ChatGPT-generated and human-annotated. For grammatical errors with clues, we proposed a heuristic method to guide ChatGPT to generate ungrammatical sentences by providing those clues. For grammatical errors without clues, we collected ungrammatical sentences from publicly available websites and manually corrected them. In addition, we employed an error-invariant augmentation method to enhance the ability of the model to correct native Chinese grammatical errors. We ultimately constructed about 1k parallel data and utilized these data to fine-tune open-source LLMs (e.g., Phoenix, released by The Chinese University of Hong Kong, Shenzhen) with instruction tuning. The experimental results show that GrammarGPT outperforms the existing SOTA system significantly. Although model parameters are 20x larger than the SOTA baseline, the required amount of data for instruction tuning is 1200x smaller, illustrating the potential of open-source LLMs on native CGEC. Our GrammarGPT ranks 3 rd on NLPCC2023 SharedTask1, demonstrating our approach’s effectiveness. The code and data are available at https://***/FreedomIntelligence/GrammarGPT. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Error correction

来源：评论

学校读者我要写书评

暂无评论

Detection and Discrimination of Arabic Phonemes Using Long Short-Term Memory (LSTM) Model 11

Detection and Discrimination of Arabic Phonemes Using Long S...

引用

11th IEEE International Conference on Intelligent Computing and Information Systems, ICICIS 2023

作者： Tolba, Rahma M. Fayed, Zaki T. Alsayadi, Hamzah A. Saad, Noureldin O. Elarif, Taha Ain Shams University Faculty of Computer and Information Sciences Computer Science Department Cairo Egypt Ibb University Faculty of Sciences Computer Science Department Yemen Canadian International Collage School of Computer Science Data Science Department New Cairo Egypt

ISBN: (纸本)9798350322101

Phonetics is a crucial branch of linguistics that studies human speech sounds and is essential for language learning, speech therapy, and speech technology development. However, current Arabic speech systems cannot instantly analyze and detect mispronunciation of Arabic phonemes, only offering a recording feature for self-evaluation. So, this paper presents a state-of-the-art approach using a Long Short-Term Memory (LSTM) model for detecting and discriminating between Arabic phonemes. The model was tested on a collected dataset for the Arabic phonemes /s/ 'sin' and /s/ 'ad and demonstrated excellent performance in discriminating between them despite their close articulation. The model achieved an accuracy of 97.33% and outperformed other techniques proposed for small datasets. The model can be utilized either independently or integrated into a computer-assisted language learning (CALL) system. Furthermore, our findings suggest that there is significant room for improvement in this area, particularly in collecting large audio datasets for all the Arabic phonemes, refining the algorithms, and optimizing the data processing pipeline. © 2023 IEEE.

关键词： Brain

来源：评论

学校读者我要写书评

暂无评论

Fast Parallel Recovery for Transactional Stream Processing on Multicores 40

Fast Parallel Recovery for Transactional Stream Processing o...

引用

40th IEEE International Conference on data Engineering, ICDE 2024

作者： Zhao, Jianjun Liu, Haikun Zhang, Shuhao Duan, Zhuohui Liao, Xiaofei Jin, Hai Zhang, Yu School of Computer Science and Technology Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Wuhan China School of Computer Science and Engineering Nanyang Technological University Singapore

ISBN: (纸本)9798350317152

Transactional stream processing engines (TSPEs) have gained increasing attention due to their capability of processing real-time stream applications with transactional semantics. However, TSPEs remain susceptible to system failures and power outages. Existing TSPEs mainly focus on performance improvement, but still face a significant challenge to guarantee fault tolerance while offering high-performance services. We revisit commonly-used fault tolerance approaches in stream processing and database systems, and find that these approaches do not work well on TSPEs due to complex data dependencies. In this paper, we propose a novel TSPE called MorphStreamR to achieve fast failure recovery while guaranteeing low performance overhead at runtime. The key idea of MorphStreamR is to record intermediate results of resolved dependencies at runtime, and thus eliminate data dependencies to improve task parallelism during failure recovery. MorphStreamR further mitigates the runtime overhead by selectively tracking data dependencies and incorporating workload-aware log commitment. Experimental results show that MorphStreamR can significantly reduce the recovery time by up to 3.1 x while experiencing much less performance slowdown at runtime, compared with other applicable fault tolerance approaches. © 2024 IEEE.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Importance-Awareness Masking Network for Robust Document Retrieval

Importance-Awareness Masking Network for Robust Document Ret...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Liu, Junping He, Jiaqi Hu, Xinrong Yang, Wangli Yang, Jie Guo, Yi School of Computer Science and Artificial Intelligence Wuhan Textile University Wuhan China School of Computing and Information Technology University of Wollongong Wollongong Australia School of Computer Data and Mathematical Sciences Western Sydney University Australia

ISBN: (纸本)9798350368741

In this paper, we introduce the IMPortance-awaReness maskIng NeTwork (IMPRINT), a novel approach to enhance the robustness of document retrieval systems against query variations, particularly those containing misspellings. Unlike previous models that treat all query components (words/features) equally, IMPRINT prioritizes the most important components while masking out less relevant ones. Specifically, we propose a Mutual Information-based measure to quantify component importance and integrate it into a dynamic masking mechanism that adjusts the retention probability of each component. Our method is evaluated on a combination of three benchmark datasets and three types of query variations. The experimental results show substantial performance gains compared to state-of-the-art models, achieving an average improvement of 1.2 absolute MRR@10 points in retrieval accuracy. © 2025 IEEE.

关键词： Document Retrieval Model Robustness Query Variations Word/Feature Importance

来源：评论

学校读者我要写书评

暂无评论

Deep Ensemble Learning Approach for Face Anti-Spoofing Detection based on Pre-trained Models

Deep Ensemble Learning Approach for Face Anti-Spoofing Detec...

引用

2024 IEEE International Conference on computer Vision and Machine Intelligence, CVMI 2024

作者： Vannurswamy, K. Shekar, B.H. Pilar, Bharathi Karunakar Kotegar, A. Jiang, Frank Mangalore University Dept of Computer Science Mangalore India Mangalore University Konaje Dept of Computer Science Mangalore India University College Mangalore Dept of Computer Science Mangalore India MAHE Manipal Dept of Data Science Manipal India Deakin University School of Information Tech Melbourne Australia

ISBN: (纸本)9798350376876

In this paper, we propose a Weighted Deep Ensemble Learning (WDEL) to increase the overall accuracy of face anti-spoofing model by exploiting multiple learning architectures. Current anti-spoofing models based on one deep learning architecture are not able to extract all important information for discriminating real and fake faces. We therefore design a framework that fuses the information of five pre-trained deep learning models, Inception, Xception, VGG16, ResNet50 and MobileNet to create a high-quality feature representation against spoofing attacks. Moreover, we apply a weighted voting mechanism with each weight being reliant on the corresponding classifier's performance from real and spoof classification considering different metrics: precision for the non-spoof classification and recall for the spoof classification. The final prediction is obtained by aggregating these multiple classifiers using their optimal weights in an ensemble fashion. Additionally, we provide an experiment comparing the performance of these aforementioned pre-trained deep learning models. The comparison reveals that MobileNet, Xception, and ResNet50 exhibit the highest recall for spoofing and precision for real images, along with improved F1 score values in detection. This establishes them as optimal choices for facial spoofing detection. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Provably Secure Symmetric-Key Privacy-Preserving authenticated key agreement protocol 13

Provably Secure Symmetric-Key Privacy-Preserving authenticat...

引用

13th International Conference on Communication and Network Security, ICCNS 2023

作者： Ni, Liang Zhou, Hengsheng Zhang, Yawei Liu, Xiaoyan Gu, Bingke School of Computer Science Henan Development and Innovation Laboratory of Industrial Internet Security Big Data Zhongyuan University of Technology Henan Zhengzhou China School of Computer Science Fujian Normal University Fujian China School of Computer Science Zhongyuan University of Technology Henan Zhengzhou China

ISBN: (纸本)9798400707964

In recent years, the rapid development of Quantum Computing technology has had a huge impact on the security of traditional Public Key Cryptography. However, the security of Hash functions and Symmetric Cryptographic algorithms is relatively less impacted by this security threaten and commonly have higher execution efficiency. This two primitives can offset the security impact by appropriately increasing the output scale of a Hash function and symmetric key respectively. Therefore, the Symmetric Key Authenticated Key Agreement(SAKA) protocol which constructed mainly by Hash functions and symmetric cryptographic algorithms has gradually attracted more attention. In this paper we proposed a provably secure SAKA scheme. We extend the security model used by Ferreira et al to formally prove the security of our protocol by introducing a criterion for indistinguishability of identities and Synchronization Robustness property. Our protocol not only captures basic security properties such as Perfect Forward Secrecy, Entity Authentication and Key Indistinguishability but also achieved Privacy Preservation and Full Synchronization Robustness. Furthermore, its implementation does not need to rely on a huge and complex Public key Infrastructure(PKI), so we can easily deploy it in a coucurrent environment. Comparison results indicate that the proposal consumes the least computation resource and reasonable communication cost among nine SAKA protocols while equiping 10 security properties. © 2023 ACM.

关键词： Hash functions

来源：评论

学校读者我要写书评

暂无评论

Image Inpainting Detection Based on High-Pass Filter Attention Network

引用

computer Systems science & Engineering 2022年第12期43卷 1145-1154页

作者： Can Xiao Feng Li Dengyong Zhang Pu Huang Xiangling Ding Victor S.Sheng Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha University of Science and TechnologyChangsha410114China School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha410114China School of Computer Science and Engineering Hunan University of Science and TechnologyXiangtan411004China Department of Computer Science Texas Tech UniversityLubbock79409USA

Image inpainting based on deep learning has been greatly *** original purpose of image inpainting was to repair some broken photos, suchas inpainting artifacts. However, it may also be used for malicious operations,such as destroying evidence. Therefore, detection and localization of imageinpainting operations are essential. Recent research shows that high-pass filteringfull convolutional network (HPFCN) is applied to image inpainting detection andachieves good results. However, those methods did not consider the spatial location and channel information of the feature map. To solve these shortcomings, weintroduce the squeezed excitation blocks (SE) and propose a high-pass filter attention full convolutional network (HPACN). In feature extraction, we apply concurrent spatial and channel attention (scSE) to enhance feature extraction and obtainmore information. Channel attention (cSE) is introduced in upsampling toenhance detection and localization. The experimental results show that the proposed method can achieve improvement on ImageNet.

关键词： Image inpainting detection spatial attention channel attention full convolutional network high-pass filter

来源：评论

学校读者我要写书评

暂无评论

data-Driven Design for Anomaly Detection in Network Access Control Systems 2

Data-Driven Design for Anomaly Detection in Network Access C...

引用

2nd International Conference on Business Analytics for Technology and Security, ICBATS 2023

作者： Muhammad, Musa Abubakar Caraffini, Fabio Fasanmade, Adebamigbe Ishola, Olabayo Mohammed, Kabiru Morden, Jarrad Coventry University School of Computing Mathematics and Data Science Coventry United Kingdom Swansea University Department of Computer Science Swansea United Kingdom De Montfort University School of Computer Science and Informatics Leicester United Kingdom

ISBN: (纸本)9798350335644

Current network access control systems can contain unpredictable interactions between multiple device models, multiple network protocol layers (e.g. TCP, UDP and ICMP), hardware, and clock-skew-specific influences, and cannot detect or identify abnormal behaviours based on the type of device. To complicate things further, the 'bring your own device' policy is increasing security threats, vulnerabilities, and risks to enterprise network environments, making intrusion detection and prevention systems unable to detect illegal and unauthorised access to devices in the enterprise network. The consequences can be disastrous. In this light, this work propose a simple but effective clustering approach capable of separating normal and abnormal network traffic patterns to detect such challenges (anomalies). The work in this paper applies this approach to single devices and aggregations of data per device type. Additionally, this work propose plotting the notched box for each cluster to acquire a better understanding of their data distributions and measuring the clusters' performance using the Adjusted Rand Index. The results show that the proposed method is valid, can be used in several contexts, and features a 95% confidence that most single device and device type distributions overlap, which makes them equivalently usable for anomaly detection purposes. © 2023 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

CONSTRUCTURE: Benchmarking CONcept STRUCTUre REasoning for Multimodal Large Language Models

CONSTRUCTURE: Benchmarking CONcept STRUCTUre REasoning for M...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Zha, Zhiwei Zhu, Xiangru Xu, Yuanyi Huang, Chenghua Liu, Jingping Li, Zhixu Wang, Xuwu Xiao, Yanghua Yang, Bei Xu, XiaoXiao Shanghai Key Laboratory of Data Science School of Computer Science Fudan University China School of Information Science and Engineering East China University of Science and Technology China Alibaba Group China School of Information Renmin University of China China Renmin University of China China

ISBN: (纸本)9798891761681

Multimodal Large Language Models (MLLMs) have shown promising results in various tasks, but their ability to perceive the visual world with deep, hierarchical understanding similar to humans remains uncertain. To address this gap, we introduce CONSTRUCTURE, a novel concept-level benchmark to assess MLLMs' hierarchical concept understanding and reasoning abilities. Our goal is to evaluate MLLMs across four key aspects: 1) Understanding atomic concepts at different levels of abstraction;2) Performing upward abstraction reasoning across concepts;3) Achieving downward concretization reasoning across concepts;and 4) Conducting multi-hop reasoning between sibling or common ancestor concepts. Our findings indicate that even state-of-the-art multimodal models struggle with concept structure reasoning (e.g., GPT-4o averages a score of 62.1%). We summarize key findings of MLLMs in concept structure reasoning evaluation. Morever, we provide key insights from experiments using CoT prompting and fine-tuning to enhance their abilities. © 2024 Association for Computational Linguistics.

关键词： Abstracting

来源：评论

学校读者我要写书评

暂无评论

BATON: symphony of random testing and concolic testing through machine learning and taint analysis

引用

science China(Information sciences) 2023年第3期66卷 97-117页

作者： Bihuan CHEN Yang LIU Xin PENG Yijian WU Shengchao QIN School of Computer Science Fudan University Shanghai Key Laboratory of Data Science Fudan University School of Computer Science and Engineering Nanyang Technological University School of Computing Teesside University

Random testing is scalable but often fails to hit corner program behaviors,while systematic testing （e.g.,concolic execution） is promising to cover corner program behaviors but is not scalable to explore all program *** attempts to integrate random testing with systematic testing lack targeted *** this paper,we propose a guided hybrid testing approach,named BATON,to synergize random testing with concolic *** integrates the knowledge inside test cases and their executions into a conditional execution graph,and uses such knowledge to guide test case ***,we learn classification models for some conditionals in the conditional execution graph in a demand-driven *** models are used to guide random testing to reach and cover partially-covered *** further employ targeted concolic testing to cover conditionals that cannot be fully covered by guided random *** implemented BATONfor Java and evaluated it on three *** results show that BATONimproved branch coverage and mutation score over random testing by 16.2%–29.4%and 19.0%–30.0%,over adaptive random testing by 16.8%–33.8%and 19.4%–34.2%,over concolic testing by 2.3%–29.9%and 2.9%–30.1%,and over simple hybrid testing by 1.6%–14.5%and 1.4%–18.7%.

关键词： system testing random testing concolic testing

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：