检索结果-内蒙古大学图书馆

IEEE International Requirements Engineering Conference Workshops (REW)

作者： Kun Ruan Xiaohong Chen Zhi Jin Shanghai Key Laboratory of Trustworthy Computing East China Normal University Shanghai China School of Computer Science Peking University China Ministry of Education Key Lab. of High-Confidence of Software Technologies (PKU) China

Requirements modeling is a crucial tool for requirements analysis and has been demonstrated to aid in the comprehension and analysis of requirements. However, constructing requirements models from natural language descriptions in user requirements documents can be a time-consuming task. With the growing attention given to large language models, they have become an integral component of natural language processing. Consequently, utilizing large language models to facilitate the construction of high-quality requirement models is appealing. This paper presents an automated framework for requirement model generation that incorporates ChatGPT-based zero-shot learning to extract requirement models from requirement texts and subsequently compose them using predefined rules. The framework defines the requirement extraction task of ChatGPT by designing appropriate prompt, and it generates requirement models by employing composition rules. Furthermore, a case study on a digital home system is conducted to validate the feasibility of the framework in assisting requirements modeling.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An LLM-Empowered Adaptive Evolutionary Algorithm For Multi-Component Deep Learning systems

arXiv

引用

arXiv 2025年

作者： Tian, Haoxiang Han, Xingshuo Wu, Guoquan Guo, An Zhou, Yuan Zhang, Jie Li, Shuo Wei, Jun Zhang, Tianwei Key Lab of System Software CAS State Key Lab of Computer Science ISCAS Beijing China University of CAS Beijing China Nanyang Technological University Singapore Continental-NTU Corporate Lab Singapore Nanjing Institute of Software Technology University of CAS Nanjing China Nanjing University Nanjing China Zhejiang Sci-Tech University Hangzhou China CFAR IHPC A*STAR Singapore

Multi-objective evolutionary algorithms (MOEAs) are widely used for searching optimal solutions in complex multi-component applications. Traditional MOEAs for multi-component deep learning (MCDL) systems face challenges in enhancing the search efficiency while maintaining the diversity. To combat these, this paper proposes µMOEA, the first LLM-empowered adaptive evolutionary search algorithm to detect safety violations in MCDL systems. Inspired by the context-understanding ability of Large Language Models (LLMs), µMOEA promotes the LLM to comprehend the optimization problem and generate an initial population tailed to evolutionary objectives. Subsequently, it employs adaptive selection and variation to iteratively produce offspring, balancing the evolutionary efficiency and diversity. During the evolutionary process, to navigate away from the local optima, µMOEA integrates the evolutionary experience back into the LLM. This utilization harnesses the LLM's quantitative reasoning prowess to generate differential seeds, breaking away from current optimal solutions. We evaluate µMOEA in finding safety violations of MCDL systems, and compare its performance with state-of-the-art MOEA methods. Experimental results show that µMOEA can significantly improve the efficiency and diversity of the evolutionary search. © 2025, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples

arXiv

引用

arXiv 2024年

作者： Zhou, Ziqi Li, Minghui Liu, Wei Hu, Shengshan Zhang, Yechao Wan, Wei Xue, Lulu Zhang, Leo Yu Yao, Dezhong Jin, Hai National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab. China Cluster and Grid Computing Lab. China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Computer Science and Technology Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

With the evolution of self-supervised learning, the pre-training paradigm has emerged as a predominant solution within the deep learning landscape. Model providers furnish pre-trained encoders designed to function as versatile feature extractors, enabling downstream users to harness the benefits of expansive models with minimal effort through fine-tuning. Nevertheless, recent works have exposed a vulnerability in pre-trained encoders, highlighting their susceptibility to downstream-agnostic adversarial examples (DAEs) meticulously crafted by attackers. The lingering question pertains to the feasibility of fortifying the robustness of downstream models against DAEs, particularly in scenarios where the pre-trained encoders are publicly accessible to the attackers. In this paper, we initially delve into existing defensive mechanisms against adversarial examples within the pre-training paradigm. Our findings reveal that the failure of current defenses stems from the domain shift between pre-training data and downstream tasks, as well as the sensitivity of encoder parameters. In response to these challenges, we propose Genetic Evolution-Nurtured Adversarial Fine-tuning (Gen-AF), a two-stage adversarial fine-tuning approach aimed at enhancing the robustness of downstream models. Gen-AF employs a genetic-directed dual-track adversarial fine-tuning strategy in its first stage to effectively inherit the pre-trained encoder. This involves optimizing the pre-trained encoder and classifier separately while incorporating genetic regularization to preserve the model’s topology. In the second stage, Gen-AF assesses the robust sensitivity of each layer and creates a dictionary, based on which the top-k robust redundant layers are selected with the remaining layers held fixed. Upon this foundation, we conduct evolutionary adaptability fine-tuning to further enhance the model’s generalizability. Our extensive experiments, conducted across ten self-supervised training methods and six

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Intersecting-boundary-sensitive fingerprinting for tampering detection of DNN models 24

Intersecting-boundary-sensitive fingerprinting for tampering...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Xiaofan Bai Chaoxiang He Xiaojing Ma Bin Benjamin Zhu Hai Jin School of Cyber Science and Engineering Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System and Services Computing Technology and System Lab and Hubei Engineering Research Center on Big Data Security and Hubei Key Laboratory of Distributed System Security Microsoft School of Computer Science and Technology Huazhong University of Science and Technology and National Engineering Research Center for Big Data Technology and System and Services Computing Technology and System Lab and Cluster and Grid Computing Lab.

Cloud-based AI services offer numerous benefits but also introduce vulnerabilities, allowing for tampering with deployed DNN models, ranging from injecting malicious behaviors to reducing computing resources. Fingerprint samples are generated to query models to detect such tampering. In this paper, we present Intersecting-Boundary-Sensitive Fingerprinting (IBSF), a novel method for black-box integrity verification of DNN models using only top-1 lab.ls. Recognizing that tampering with a model alters its decision boundary, IBSF crafts fingerprint samples from normal samples by maximizing the partial Shannon entropy of a selected subset of categories to position the fingerprint samples near decision boundaries where the categories in the subset intersect. These fingerprint samples are almost indistinguishable from their source samples. We theoretically establish and confirm experimentally that these fingerprint samples' expected sensitivity to tampering increases with the cardinality of the subset. Extensive evaluation demonstrates that IBSF surpasses existing state-of-the-art fingerprinting methods, particularly with larger subset cardinality, establishing its state-of-the-art performance in black-box tampering detection using only top-1 lab.ls. The IBSF code is availab.e at: https://***/CGCL-codes/IBSF.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Bridge the Gap between Supervised and Unsupervised Learning for Fine-Grained Classification

SSRN

引用

SSRN 2023年

作者： Wang, Jiabao Li, Yang Wei, Xiu-Shen Li, Hang Miao, Zhuang Zhang, Rui Army Engineering University of PLA Jiangsu Nanjing210007 China PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Nanjing210094 China State Key Lab. for Novel Software Technology Nanjing University Jiangsu Nanjing210023 China

Unsupervised learning technology has caught up with or even surpassed supervised learning technology in general object classification (GOC) and person re-identification (re-ID). However, it is found that the unsupervised learning of fine-grained visual classification (FGVC) is more challenging than GOC and person re-ID. In order to bridge the gap between unsupervised and supervised learning for FGVC, we investigate the essential factors for the performance gap. Furthermore, we propose a simple, effective, and practical method, termed as UFCL, to alleviate the gap. Three key issues are concerned and improved: First, we introduce a robust and powerful backbone, ResNet50-IBN, which has an ability of domain adaptation when we transfer ImageNet pre-trained models to FGVC tasks. Next, we propose to introduce HDBSCAN instead of DBSCAN to do clustering, which can generate better clusters for adjacent categories with fewer hyper-parameters. Finally, we propose a weighted feature agent and its updating mechanism to do contrastive learning, which can improve the optimization process of learning the parameters of the network. The effectiveness of our UFCL is verified on four public FGVC datasets. Under the unsupervised FGVC setting, we achieve state-of-the-art results, and analyze the key factors and the important parameters to provide a practical guidance. © 2023, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

A mathematical model for efficient extraction of key locations from point-cloud data in track area

引用

Industrial Artificial Intelligence 2023年第1期1卷 1-14页

作者： Chen, Shuyue Wu, Jiaolv Lu, Jian Wang, Xizhao College of Mathematics and Statistics Shenzhen University Shenzhen China School of Software Engineering Shenzhen Institue of Information Technology Shenzhen China Shenzhen No. 3 Vocational School of Technology Shenzhen China College of Engineering Huaqiao University Quanzhou China Shenzhen Key Lab. of Advanced Machine Learning and Applications Shenzhen University Shenzhen China Guangdong Key Lab. of Intelligent Information Process Shenzhen University Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China

During the construction of a metro system, it is inevitable that deviations will occur between the excavated tunnel and the original designed scheme. As such, it is necessary to adjust the designed scheme to accommodate these discrepancies. Specifically, the adjustment of the designed scheme involves a rigorous process of repeatedly selecting and verifying the feasibility of the proposed modifications using point-cloud data obtained from the tunnel. However, this process can be considerably time-consuming due to the large-scale and potentially redundant nature of the point-cloud data. This paper proposes a mathematical model for point-cloud data acquired in measuring a mined tunnel, which may deviate from the originally designed one. The modeling, which mainly includes determining its normal plane, and building the equation of tunnel point-cloud data, is to quickly extract several key locations in the tunnel surface for modifying the original design in order to achieve a minimum error between the modified design and the mined tunnel. In comparison with the conventional processing of extracting several key locations directly from point-cloud data, our model shows a significant promotion of extraction efficiency under an acceptable error bound. The model is tested in a real tunnel point-cloud data and the testing results confirm the increase of fitting accuracy and the decrease of computational load.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Efficient Workflow Scheduling Over Yarn Cluster Using Deep Reinforcement Learning

Towards Efficient Workflow Scheduling Over Yarn Cluster Usin...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Jianguo Xue Ting Wang Puyu Cai MoE Engineering Research Center of Software/Hardware Co-design Technology and Application Shanghai Key Lab. of Trustworthy Computing East China Normal University China Department of Computer Science and Engineering Michigan State University USA

Hadoop Yarn is an open-source cluster manager responsible for resource management and job scheduling. However, data-driven applications are typically organized into workflows that consist of a series of jobs with dependencies. Yarn does not manage users' workflows and only considers the current job rather than the entire workflow when scheduling. In practice, multiple workflows share the same Yarn cluster and are pre-assigned separate Yarn resource queues to avoid mutual interference. However, this coarse-grained resource division can sometimes result in low resource utilization and increased pending time of jobs on the Yarn queue. For instance, one resource queue may have exhausted its quota while still having pending jobs, while other queues may have availab.e resources but cannot begin executing any jobs due to unfulfilled data dependencies. To address this problem, we propose a deep reinforcement learning-based workflow scheduling scheme that takes into account job dependencies, job priorities, and dynamic resource usage. The proposed approach can intelligently identify and utilize free windows of different resource queues. Our simulation results demonstrate that the proposed DRL-based workflow scheduling scheme can significantly reduce the average job latency compared to existing approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SliceProp: A Slice-Wise Bidirectional Propagation Model for Interactive 3D Medical Image Segmentation 1

SliceProp: A Slice-Wise Bidirectional Propagation Model for ...

引用

1st IEEE International Conference on Medical Artificial Intelligence, MedAI 2023

作者： Xu, Xin Lu, Wenjing Lei, Jiahao Qiu, Peng Shen, Hong-Bin Yang, Yang Shanghai Jiao Tong University Key Lab. of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Department of Computer Science and Engineering Shanghai200240 China Shanghai Ninth People's Hospital Shanghai Jiao Tong University School of Medicine Department of Vascular Surgery China Shanghai Jiao Tong University Institute of Image Processing and Pattern Recognition Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China

ISBN: (纸本)9798350358780

Interactive medical image segmentation methods have become increasingly popular in recent years. These methods combine manual lab.ling and automatic segmentation, reducing the workload of annotation while maintaining high accuracy. However, most current interactive segmentation frameworks are limited to 2D image data, and are not suitable for 3D image data due to the large size and high complexity of 3D data, as well as the challenges posed by information asymmetry and sparse annotation. In this paper, we propose SliceProp, an interactive segmentation framework that implements slice-wise lab.l Bidirectional Propagation (LBP) for 3D medical image segmentation. SliceProp extends the interactive 2D image segmentation algorithm to 3D image segmentation, and can handle 3D data with large size and high complexity. Moreover, equipped with a Backtracking Feedback Check (BFC) module, SliceProp effectively addresses the issues of information asymmetry and spatial sparse annotation in 3D medical image segmentation. Additionally, we adopt an uncertainty-based criterion to pri-oritize the slices to be refined interactively, which enhances the efficiency of the interaction process by enabling the model to focus on the regions with the most unreliable predictions. SliceProp is evaluated on two datasets and achieves promising results compared to state-of-the-art methods. © 2023 IEEE.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration

DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral M...

引用

International Conference on computer Vision (ICCV)

作者： Yuchun Miao Lefei Zhang Liangpei Zhang Dacheng Tao National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Hubei Luojia Laboratory State Key Lab. of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Sydney AI Centre School of Computer Science The University of Sydney

Diffusion models have recently received a surge of interest due to their impressive performance for image restoration, especially in terms of noise robustness. However, existing diffusion-based methods are trained on a large amount of training data and perform very well in-distribution, but can be quite susceptible to distribution shift. This is especially inappropriate for data-starved hyperspectral image (HSI) restoration. To tackle this problem, this work puts forth a self-supervised diffusion model for HSI restoration, namely Denoising Diffusion Spatio-Spectral Model (DDS2M), which works by inferring the parameters of the proposed Variational Spatio-Spectral Module (VS2M) during the reverse diffusion process, solely using the degraded HSI without any extra training data. In VS2M, a variational inference-based loss function is customized to enable the untrained spatial and spectral networks to learn the posterior distribution, which serves as the transitions of the sampling chain to help reverse the diffusion process. Benefiting from its self-supervised nature and the diffusion process, DDS2M enjoys stronger generalization ability to various HSIs compared to existing diffusion-based methods and superior robustness to noise compared to existing HSI restoration methods. Extensive experiments on HSI denoising, noisy HSI completion and super-resolution on a variety of HSIs demonstrate DDS2M’s superiority over the existing task-specific state-of-the-arts. Code is availab.e at: https://***/miaoyuchun/DDS2M.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Embedding Path Order for Lambda-Free Higher-Order Terms

Journal of Applied Logics

引用

Journal of Applied Logics 2021年第10期8卷 2447-2470页

作者： Bentkamp, Alexander Vrije Universiteit Amsterdam Netherlands State Key Lab. of Computer Science Institute of Software CAS Beijing China

The embedding path order, introduced in this article, is a variant of the recursive path order (RPO) for untyped λ-free higher-order terms (also called applicative first-order terms). Unlike other higher-order variants of RPO, it is a ground-total and well-founded simplification order, making it more suitable for the superposition calculus. I formally proved the order’s theoretical properties in Isabelle/HOL and evaluated the order in a prototype based on the superposition prover Zipperposition. © 2021, College Publications. All rights reserved.

关键词： Embeddings

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：