检索结果-内蒙古大学图书馆

Improving Adversarial Transferability of Vision-Language Pre-training Models through Collaborative Multimodal Interaction

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Fu, Jiyuan Chen, Zhaoyu Jiang, Kaixun Guo, Haijing Wang, Jiafeng Gao, Shuyong Zhang, Wenqiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai China Engineering Research Center of AI & Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai China

Despite the substantial advancements in Vision-Language Pre-training (VLP) models, their susceptibility to adversarial attacks poses a significant challenge. Existing work rarely studies the transferability of attacks on VLP models, resulting in a substantial performance gap from white-box attacks. We observe that prior work overlooks the interaction mechanisms between modalities, which plays a crucial role in understanding the intricacies of VLP models. In response, we propose a novel attack, called Collaborative Multimodal Interaction Attack (CMI-Attack), leveraging modality interaction through embedding guidance and interaction enhancement. Specifically, attacking text at the embedding level while preserving semantics, as well as utilizing interaction image gradients to enhance constraints on perturbations of texts and images. Significantly, in the image-text retrieval task on Flickr30K dataset, CMI-Attack raises the transfer success rates from ALBEF to TCL, CLIPViT and CLIPCNN by 8.11%-16.75% over state-of-the-art methods. Moreover, CMI-Attack also demonstrates superior performance in cross-task generalization scenarios. Our work addresses the underexplored realm of transfer attacks on VLP models, shedding light on the importance of modality interaction for enhanced adversarial robustness. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

Enhancement of Antenna Gain With M&W Slot Model on EBG Reflector Structure Utilizing Reflection Phase Diagram Analysis

学校读者我要写书评

暂无评论

Enhancement of Antenna Gain With M&W Slot Model on EBG Refle...

International Electrical engineering Congress (iEECON)

作者： Payu Chakunna Nopphadon Phusam Pisit Mahatthanajatuphat Wilaiporn Lee Thanakarn Suangun School of Engineering University of Phayao Phayao Thailand Regular Program in Mathematics-Science Yothinburana School Bangkok Thailand Department of Electrical and Computer Engineering King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798331543952

ISBN: (纸本)9798331543969

This research demonstrates the improvement in antenna gain by utilizing an EBG reflector. The EBG reflector includes a unit cell with grooves shaped like the letters M and W to mitigate surface waves at a frequency of 2.45 GHz. The suggested EBG reflector is constructed on an FR-4 substrate exhibiting a uniform dielectric constant of 4.3, a loss tangent of 0.025, and a thickness of 1.6 mm. Furthermore, the impact of boosting the antenna gain can be analyzed by the reflection phase diagram, which exhibits values ranging from +90° to -90° at the resonant frequency of the reflector. The proposed EBG reflector comprises 5x5 unit cells. The testing results of the reflector's performance, when integrated with a general dipole antenna, indicate that it can effectively respond to the resonant frequency of 2.45 GHz, with a |S 11 | level below -10 dB within the frequency range of 2.2 GHz to 2.6 GHz, thereby adequately covering the IEEE 802.11 b/g/n band. The suggested EBG reflector exhibits a gain of around 6.5 dBi at 2.45 GHz, indicating its efficacy in enhancing gain relative to the standard PEC reflector. Moreover, the radiation pattern is also directional.

关键词： Slot antennas Dipole antennas Resonant frequency Reflector antennas Metamaterials Reflection Gain Standards Periodic structures Antenna radiation patterns

Implementation and Manufacturing of DT Sensor Ecosystem for Real-Time Monitoring of Virtual 3D Printers

学校读者我要写书评

暂无评论

SN computer science 2023年第5期4卷 556页

作者： Reddy, K. Shyam Sunder Rajesh, R. Raj, P. Anantha Christu Arya, Namita Bhaskaran, R. Prasad, J. Laxmi Department of Information Technology Vasavi College of Engineering Hyderabad India School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Chennai India Department of Robotics Engineering Karunya Institute of Technology and Sciences Karunya Nagar Tamil Nadu Coimbatore India Galgotias University Gautam Buddha Nagar Uttar Pradesh Greater Noida India Department of IT PSNA College of Engineering and Technology Tamil Nadu Dindigul India Department of Mechanical Engineering MLR Institute of Technology Telangana Hyderabad India

A full-featured digital representation of all the characteristics of a physical system or object is known as a Digital Twin (DT) and smart manufacturing is made possible in large part by our relationship to the environment. A DT system is simulated in real time and with high accuracy. While, maintaining constant synchronization provides a comprehensive system with physical system management. The use of a DT for a 3D printer is covered in this article which presents the 3D printer’s settings and relevant warnings in real time and uses a lightweight Augmented Reality (AR) model to let users control the basic printer functions that encourage communication between two parties. The tri-model simulates actual physical behavior while operating simultaneously and digital model features. To verify the architecture and method that have been proposed, further investigation is done using a case study of a DT facility using open-source 3D printing. Additionally, future research and conclusions are highlighted to offer illuminating information finally to the two industries and the academic world. We have an economical DT ecosystem, robust and reproducible, thereby allowing the addition of DT capabilities to legacy equipment;creating analytics using historical data that have been collected. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： 3D printing Cost-efficient Digital model DT Smart manufacturing

From Efficient Multimodal Models to World Models: A Survey

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Mai, Xinji Tao, Zeng Lin, Junxiong Wang, Haoran Chang, Yang Kang, Yanlan Wang, Yan Zhang, Wenqiang Shanghai Engineering Research Center of AI and Robotics Academy for Engineering and Technology Fudan University Shanghai China Engineering Research Center of AI and Robotics Ministry of Education Academy for Engineering and Technology Fudan University Shanghai China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China

Multimodal Large Models (MLMs) are becoming a significant research focus, combining powerful large language models with multimodal learning to perform complex tasks across different data modalities. This review explores the latest developments and challenges in MLMs, emphasizing their potential in achieving artificial general intelligence and as a pathway to world models. We provide an overview of key techniques such as Multimodal Chain of Thought (M-COT), Multimodal Instruction Tuning (M-IT), and Multimodal In-Context Learning (M-ICL). Additionally, we discuss both the fundamental and specific technologies of multimodal models, highlighting their applications, input/output modalities, and design characteristics. Despite significant advancements, the development of a unified multimodal model remains elusive. We discuss the integration of 3D generation and embodied intelligence to enhance world simulation capabilities and propose incorporating external rule systems for improved reasoning and decision-making. Finally, we outline future research directions to address these challenges and advance the field. Copyright © 2024, The Authors. All rights reserved.

关键词： Decision making

Hotel Information System Management Audit Using COBIT 2019 - APO12

学校读者我要写书评

暂无评论

Hotel Information System Management Audit Using COBIT 2019 -...

Innovation in Technology (INOCON), IEEE International Conference for

作者： Joy Nashar Utamajaya Suhono Harso Supangat Ford Lumban Gaol Benny Ranti BINUS Graduate Program School of Information System BINUS University Jakarta Indonesia School of Electrical Engineering and Informatic Bandung Institute of Technology Bandung Indonesia Computer Science Department BINUS Graduate Program Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department Faculty of Computer Science University of Indonesia Depok Indonesia

In general, public or private organizations or companies have used information-based technology as a support to improve business performance to be more effective and efficient in order to achieve a company’s business goals. Given the large contribution of information technology in the application of hotel applications as a supporting information system, it is a vital system that must avoid risks that can hinder and cause harm to the hotel management business processes. Therefore, it is necessary to carry out risk management using the COBIT 5 framework to manage possible risks that may occur based on the APO12 (Manage Risk) domain. From the evaluation results of the data obtained through observation and interviews as well as the calculation of the results of the questionnaire based on 6 APO12 subdomains, the results of the risk management level capability assessment in hotel applications are still at level 3 or have reached the level of established process with the target to be achieved at level 4 resulting in a gap of 1 level. Based on the results obtained, it is necessary to propose recommendations that can be used by hotels in improving the application of information technology risk management so that in the future it can achieve the expected target level so that the APO12 level of capability can increase and become more optimal.

关键词： Technological innovation Companies Risk management Information technology Interviews Information systems

ROBUST: 221 Bugs in the Robot Operating System

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Timperley, Christopher S. van der Hoorn, Gijs Santos, André Deshpande, Harshavardhan Wąsowski, Andrzej School of Computer Science Carnegie Mellon University United States Department of Cognitive Robotics faculty of Mechanical Maritime and Materials Engineering Delft University of Technology Netherlands VORTEX CoLab Portugal Fraunhofer Institute for Manufacturing Engineering and Automation IPA Germany Department of Computer Science IT University of Copenhagen Denmark

As robotic systems such as autonomous cars and delivery drones assume greater roles and responsibilities within society, the likelihood and impact of catastrophic software failure within those systems is increased. To aid researchers in the development of new methods to measure and assure the safety and quality of robotics software, we systematically curated a dataset of 221 bugs across 7 popular and diverse software systems implemented via the Robot Operating System (ROS). We produce historically accurate recreations of each of the 221 defective software versions in the form of Docker images, and use a grounded theory approach to examine and categorize their corresponding faults, failures, and fixes. Finally, we reflect on the implications of our findings and outline future research directions for the community. © 2024, CC BY-NC-SA.

关键词： Robot Operating System

AI-Powered Precision: Revolutionizing Skin Cancer Detection with Advanced Machine Learning

学校读者我要写书评

暂无评论

AI-Powered Precision: Revolutionizing Skin Cancer Detection ...

Novel Intelligent and Leading Emerging sciences Conference (NILES)

作者： Mariam M. Abd-elaziz Basma M. Elshafie Hagar Abdullah Karen Y. Gerges Ingy A. Ghanem Khaled M. Fouad Amir El-Ghamry Ibrahim F. Moawad Sara El-Metwally Faculty of computer science and engineering New Mansoura university Dakahlia Egypt Faculty of Engineering Computer and Communication program Faculty of Computer Science and Engineering Mansoura University New Mansoura University Information Systems Faculty of Computers and Artificial Intelligence Benha University Egypt Faculty of Computer and Information Sciences Faculty of Computer Science and Engineering Mansoura University New Mansoura University Egypt School of Engineering and Computer Science University of Hertfordshire GAF Egypt Faculty of Computer and Information Sciences Ain Shams University Cairo Egypt Department of Artificial Intelligence College of Computer Science and Engineering Taibah University Medina Saudi Arabia

ISBN: (数字)9798350378511

ISBN: (纸本)9798350378528

Skin cancer's increasing incidence rates necessitate advanced diagnostic tools. This research uses MobileNet architecture to develop an enhanced system for skin cancer detection. MobileNet's efficient CNN architecture, combined with Keras for model lifecycle management and TensorFlow for real-time inference, forms the foundation of this study. Using 10,015 dermatoscopic pictures from seven different classifications of skin cancer in the HAM10000 dataset, the methodology includes several key stages: image preprocessing to correct illumination and adjust resolution, data augmentation to balance the dataset, and model training using Transfer Learning. The MobileNet model was trained over 50 epochs with a comprehensive architecture incorporating multiple specialized layers. By combining large-scale data analysis and adaptive learning, this methodology demonstrates The revolutionary possibilities of AI and ML in improving skin cancer diagnostics and public health outcomes. Keywords Skin Cancer Detection, Dermatoscopic Images, Machine Learning, Deep Learning

关键词： Training Deep learning Adaptation models Image resolution Transfer learning Lighting Real-time systems Image preprocessing Public healthcare Skin cancer

An Extended Convergence Result for Behaviour Tree Controllers

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Sprague, Christopher Iliffe Ögren, Petter The Robotics Perception and Learning Lab. School of Electrical Engineering and Computer Science Royal Institute of Technology KTH StockholmSE-100 44 Sweden

Behavior trees (BTs) are an optimally modular framework to assemble hierarchical hybrid control policies from a set of low-level control policies using a tree structure. Many robotic tasks are naturally decomposed into a hierarchy of control tasks, and modularity is a well-known tool for handling complexity, therefor behavior trees have garnered widespread usage in the robotics community. In this paper, we study the convergence of BTs, in the sense of reaching a desired part of the state space. Earlier results on BT convergence were often tailored to specific families of BTs, created using different design principles. The results of this paper generalize the earlier results and also include new cases of cyclic switching not covered in the literature. Copyright © 2023, The Authors. All rights reserved.

关键词： Trees (mathematics)

Contrast-augmented Diffusion Model with Fine-grained Sequence Alignment for Markup-to-Image Generation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhong, Guojin Yuan, Jin Wang, Pan Yang, Kailun Guan, Weili Li, Zhiyong College of Computer Science and Electronic Engineering Hunan University Changsha China School of Robotics Hunan University Changsha China Monash University Melbourne Australia

The recently rising markup-to-image generation poses greater challenges as compared to natural image generation, due to its low tolerance for errors as well as the complex sequence and context correlations between markup and rendered image. This paper proposes a novel model named "Contrast-augmented Diffusion Model with Fine-grained Sequence Alignment" (FSA-CDM), which introduces contrastive positive/negative samples into the diffusion model to boost performance for markup-to-image generation. Technically, we design a fine-grained cross-modal alignment module to well explore the sequence similarity between the two modalities for learning robust feature representations. To improve the generalization ability, we propose a contrast-augmented diffusion model to explicitly explore positive and negative samples by maximizing a novel contrastive variational objective, which is mathematically inferred to provide a tighter bound for the model’s optimization. Moreover, the context-aware cross attention module is developed to capture the contextual information within markup language during the denoising process, yielding better noise prediction results. Extensive experiments are conducted on four benchmark datasets from different domains, and the experimental results demonstrate the effectiveness of the proposed components in FSA-CDM, significantly exceeding state-of-the-art performance by about 2% ∼ 12% DTW improvements. The code will be released at https://***/zgj77/FSACDM. Copyright © 2023, The Authors. All rights reserved.

关键词： Alignment