检索结果-内蒙古大学图书馆

Patch is enough: naturalistic adversarial patch against vision-language pre-training models

Visual Intelligence 2024年第1期2卷 1-10页

作者： Kong, Dehong Liang, Siyuan Zhu, Xiaopeng Zhong, Yuansheng Ren, Wenqi School of Cyber Science and Technology Shenzhen Campus of Sun Yat-sen University Shenzhen 518107 China The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou 310058 China School of Computing National University of Singapore Singapore 119077 Singapore Guangdong Testing Institute of Product Quality Supervision Guangzhou 510670 China

Visual language pre-training (VLP) models have demonstrated significant success in various domains, but they remain vulnerable to adversarial attacks. Addressing these adversarial vulnerabilities is crucial for enhancing security in multi-modal learning. Traditionally, adversarial methods that target VLP models involve simultaneous perturbation of images and text. However, this approach faces significant challenges. First, adversarial perturbations often fail to translate effectively into real-world scenarios. Second, direct modifications to the text are conspicuously visible. To overcome these limitations, we propose a novel strategy that uses only image patches for attacks, thus preserving the integrity of the original text. Our method leverages prior knowledge from diffusion models to enhance the authenticity and naturalness of the perturbations. Moreover, to optimize patch placement and improve the effectiveness of our attacks, we utilize the cross-attention mechanism, which encapsulates inter-modal interactions by generating attention maps to guide strategic patch placement. Extensive experiments conducted in a white-box setting for image-to-text scenarios reveal that our proposed method significantly outperforms existing techniques, achieving a 100% attack success rate. © The Author(s) 2024.

关键词： Adversarial Patch Diffusion Model Naturalistic Physical Attack

来源：评论

学校读者我要写书评

暂无评论

MetaSLRCL: A Self-Adaptive Learning Rate and Curriculum Learning Based Framework for Few-Shot Text Classification 29

MetaSLRCL: A Self-Adaptive Learning Rate and Curriculum Lear...

引用

29th International Conference on Computational Linguistics, COLING 2022

作者： Zhao, Kailin Jin, Xiaolong Guan, Saiping Guo, Jiafeng Cheng, Xueqi CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of SciencesSchool of Computer Science and Technology University of Chinese Academy of Sciences

Due to the lack of labeled data in many realistic scenarios, a number of few-shot learning methods for text classification have been proposed, among which the meta learning based ones have recently attracted much attention. Such methods usually consist of a learner as the classifier and a meta learner for specializing the learner to different tasks. For the learner, learning rate is crucial to its performance. However, existing methods treat it as a hyper parameter and adjust it manually, which is time-consuming and laborious. Intuitively, for different tasks and neural network layers, the learning rates should be different and self-adaptive. For the meta learner, it requires a good generalization ability so as to quickly adapt to new tasks. Motivated by these issues, we propose a novel meta learning framework, called MetaSLRCL, for few-shot text classification. Specifically, we present a novel meta learning mechanism to obtain different learning rates for different tasks and neural network layers so as to enable the learner to quickly adapt to new training data. Moreover, we propose a task-oriented curriculum learning mechanism to help the meta learner achieve a better generalization ability by learning from different tasks with increasing difficulties. Extensive experiments on three benchmark datasets demonstrate the effectiveness of MetaSLRCL. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Text processing

来源：评论

学校读者我要写书评

暂无评论

Dynamic Key Generationfor Secure File Sharing System Using Time Stamp

Dynamic Key Generationfor Secure File Sharing System Using T...

引用

2024 International Conference on Innovative computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Annie Silviya, S.H. Hilda Jerlin, C.M. Mohana Suganthi, N. Helen Sathiya, V. Jayashree, R. Oviya, N. Rajalakshmi Institute of Technology Chennai India Artificial Intelligence and Data Science Panimalar Engineering College Chennai India School of Computing Computer Science and Engineering Vel Tech Rangarajan Dr Sagunthala R&d Institute of Science and Technology Chennai India Vel Tech Multi Tech Dr Rangarajan Dr Sakunthala Engineering College Chennai India

ISBN: (纸本)9798331543617

Dynamic key generation for secure file sharing system using timestamp is a novel approach to military-based secure file sharing, which adds an additional layer of security to AES-2S6 which is prominently considered as the military-grade encryption. It is designed to enhance the confidentiality and integrity of transmitted files. When a Sender uploads a file, the system prompts for a character, converts it to its ASCII value, adds the current system time value to it and utilizes this aggregated value as a unique code to secure the uploaded file. To retrieve a file, the user must possess either the file key or initiate the retrieval process during the exact time of the original upload or in the next precise clock cycle. The combination of these factors provides an extra layer of security, making unauthorized access significantly more challenging. © 2024 IEEE.

关键词： Cryptography Encryption File Sharing System Timestamp

来源：评论

学校读者我要写书评

暂无评论

data-faithful feature attribution: mitigating unobservable confounders via instrumental variables 24

Data-faithful feature attribution: mitigating unobservable c...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Qiheng Sun Haocheng Xia Jinfei Liu Zhejiang University and Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security Siebel School of Computing and Data Science University of Illinois Urbana-Champaign

ISBN: (纸本)9798331314385

The state-of-the-art feature attribution methods often neglect the influence of unobservable confounders, posing a risk of misinterpretation, especially when it is crucial for the interpretation to remain faithful to the data. To counteract this, we propose a new approach, data-faithful feature attribution, which trains a confounder-free model using instrumental variables. The cluttered effects of unobservable confounders in a model trained as such are decoupled from input features, thereby aligning the output of the model with the contribution of input features to the target feature in the data generation. Furthermore, feature attribution results produced by our method are more robust when focusing on attributions from the perspective of data generation. Our experiments on both synthetic and real-world datasets demonstrate the effectiveness of our approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IoT -Weather Integration for Enhanced Cricket Tactics with Gradient Boosting Algorithm 8

IoT -Weather Integration for Enhanced Cricket Tactics with G...

引用

8th IEEE International Conference on Computational System and Information Technology for Sustainable Solutions, CSITSS 2024

作者： Ranganathan, Chitra Sabapathy Ramesh, S. Preetha, J. Rakesh, Nitin Mohankumar, N. Sujatha, S. Mphasis Corporation Tamil Nadu Chennai India School of Computing College of Engineering and Technology Srm Institute of Science and Technology Department of Computing Technologies Tamil Nadu Chennai India Kongunadu College of Engineering and Technology Department of Artificial Intelligence & Data Science Tamil Nadu Trichy India Nagpur Campus Maharashtra Pune India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Department of Biomedical Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798331505462

From a strategic perspective, the pitch's state and the wind's direction are the most important factors in cricket. In this paper, an innovative method that uses the IoT to report the state of the pitch and the weather in real time is presented. The system gives updates on the state of the field and the weather continuously. The device can gather data on the pitch's hardness and humidity using sensors strategically placed around the field. It may also record information about the weather, including temperature, humidity, wind speed, and cloud cover. Cricket teams and players can have a deep understanding of the present playing conditions based on the real-time transmission and analysis. Teams may improve their results by making data-driven adjustments to their strategies and lineups. Teams may also be able to make more educated decisions on their approaches to batting, bowling, and fielding. The technology has the potential to alter the game's dynamics while also increasing audience participation by shedding light on cricket's many nuanced aspects. © 2024 IEEE.

关键词： Cricket (Sports)

来源：评论

学校读者我要写书评

暂无评论

RVISA: Reasoning and Verification for Implicit Sentiment Analysis

引用

IEEE Transactions on Affective computing 2025年

作者： Lai, Wenna Xie, Haoran Xu, Guandong Li, Qing Hong Kong Polytechnic University Department of Computing Hong Kong Lingnan University School of Data Science Hong Kong University of Technology Sydney School of Computer Science Data Science Institute SydneyNSW2007 Australia Education University of Hong Kong Hong Kong

Under the context of the increasing social demand for fine-grained sentiment analysis (SA), implicit sentiment analysis (ISA) poses a significant challenge owing to the absence of salient cue words in expressions. Thus, reliable reasoning is required to understand how sentiment is evoked, enabling the identification of implicit sentiments. In the era of large language models (LLMs), encoder-decoder (ED) LLMs have emerged as popular backbone models for SA applications, given their impressive text comprehension and reasoning capabilities across diverse tasks. In comparison, decoder-only (DO) LLMs exhibit superior natural language generation and in-context learning capabilities. However, their responses may contain misleading or inaccurate information. To accurately identify implicit sentiments with reliable reasoning, this study introduces a two-stage reasoning framework named Reasoning and Verification for Implicit Sentiment Analysis (RVISA), which leverages the generation ability of DO LLMs and reasoning ability of ED LLMs to train an enhanced reasoner. The framework involves three-hop reasoning prompting to explicitly furnish sentiment elements as cues. The generated rationales are then used to fine-tune an ED LLM into a skilled reasoner. Additionally, we develop a straightforward yet effective answer-based verification mechanism to ensure the reliability of reasoning learning. Evaluation of the proposed method on two benchmark datasets demonstrates that it achieves state-of-the-art performance in ISA. © 2010-2012 IEEE.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Dual-Functional Waveform Design for STAR-RIS Aided ISAC via Deep Reinforcement Learning 35

Dual-Functional Waveform Design for STAR-RIS Aided ISAC via ...

引用

35th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2024

作者： Zhang, Jifa Gong, Shiqi Lu, Weidang Xing, Chengwen Zhao, Nan Ng, Derrick Wing Kwan Niyato, Dusit Dalian University of Technology School of Information and Communication Engineering Dalian China Beijing Institute of Technology School of Cyberspace Science and Technology Beijing China Zhejiang University of Technology College of Information Engineering Hangzhou China Beijing Institute of Technology School of Information and Electronics Beijing China University of New South Wales School of Electrical Engineering and Telecommunications Sydney Australia Nanyang Technological University College of Computing and Data Science Singapore

ISBN: (纸本)9798350362244

Integrated sensing and communication (ISAC) technology effectively enables spectrum and hardware sharing between radar and communication. This paper investigates the dual-functional (DF) constant modulus waveform design for simultaneously transmitting and reconfigurable intelligent surface (STAR-RIS)-aided ISAC, in which the channel information can be used as semantic information. To investigate the performance trade-off, the weighted sum of multi-user interference (MUI) energy and waveform discrepancies is minimized via jointly optimizing the transmit waveform and the reflection and transmission coefficient matrices at STAR-RIS. Furthermore, a practical case of coupled phase shifts at STARRIS is investigated. We first formulate the optimization problem as a Markov decision process, employing a twin delayed deep deterministic policy gradient (TD3)-based deep reinforcement learning approach to address it. Simulation results verify the effectiveness of the proposed scheme. © 2024 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

HOFD: An Outdated Fact Detector for Knowledge Bases

引用

IEEE Transactions on Knowledge and data Engineering 2023年第10期35卷 10775-10789页

作者： Hao, Shuang Chai, Chengliang Li, Guoliang Tang, Nan Wang, Ning Yu, Xiang Beijing Jiaotong University School of Computer and Information Technology Beijing100044 China Beijing Key Laboratory of Traffic Data Analysis and Mining Beijing100044 China Beijing Institute of Technology School of Computer Science and Technology Beijing100811 China Beijing100190 China Qatar Computing Research Institute Hbku Doha34110 Qatar Huawei Company Gauss Department Beijing100031 China

Knowledge bases (KBs), which store high-quality information, are crucial for many applications, such as enhancing search results and serving as external sources for data cleaning. Not surprisingly, there exist outdated facts in most KBs due to the rapid change of information. Naturally, it is important to keep KBs up-to-date. Traditional wisdom has investigated the problem of using reference data (such as new facts extracted from the news) to detect outdated facts in KBs. However, existing approaches can only cover a small percentage of facts in KBs. In this paper, we propose HOFD, a novel human-in-the-loop approach for outdated fact detection in KBs. HOFD trains a binary classifier using features such as historical update frequency and update time of a fact to compute the likelihood of a fact in a KB to be outdated. Then, HOFD interacts with humans to verify whether a fact with high likelihood is indeed outdated. In addition, HOFD also uses logical rules to detect more outdated facts based on human feedback. The outdated facts detected by the logical rules will also be fed back to train the ML model further for data augmentation. Extensive experiments on real-world KBs, such as Yago and DBpedia, show the effectiveness of our solution. © 1989-2012 IEEE.

关键词： Knowledge based systems

来源：评论

学校读者我要写书评

暂无评论

STAR: Stepwise Task Augmentation and Relation Learning for Aspect Sentiment Quad Prediction

arXiv

引用

arXiv 2025年

作者： Lai, Wenna Xie, Haoran Xu, Guandong Li, Qing Department of Computing Hong Kong Polytechnic University Hong Kong School of Data Science Lingnan University Hong Kong School of Computer Science Data Science Institute University of Technology Sydney SydneyNSW Australia Department of Computing The Hong Kong Polytechnic University Hong Kong

Aspect-based sentiment analysis (ABSA) aims to identify four sentiment elements, including aspect term, aspect category, opinion term, and sentiment polarity. These elements construct the complete picture of sentiments. The most challenging task, aspect sentiment quad prediction (ASQP), predicts these elements simultaneously, hindered by difficulties in accurately coupling different sentiment elements. A key challenge is insufficient annotated data that limits the capability of models in semantic understanding and reasoning about quad prediction. To address this, we propose stepwise task augmentation and relation learning (STAR), a strategy inspired by human reasoning. STAR constructs auxiliary data to learn quadruple relationships incrementally by augmenting with pairwise and overall relation tasks derived from training data. By encouraging the model to infer causal relationships among sentiment elements without requiring additional annotations, STAR effectively enhances quad prediction. Extensive experiments demonstrate the proposed STAR exhibits superior performance on four benchmark datasets. © 2025, CC BY-NC-SA.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Black-box backdoor defense via zero-shot image purification 23

Black-box backdoor defense via zero-shot image purification

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Yucheng Shi Mengnan Du Xuansheng Wu Zihan Guan Jin Sun Ninghao Liu School of Computing University of Georgia Department of Data Science New Jersey Institute of Technology

Backdoor attacks inject poisoned samples into the training data, resulting in the misclassification of the poisoned input during a model's deployment. Defending against such attacks is challenging, especially for real-world black-box models where only query access is permitted. In this paper, we propose a novel defense framework against backdoor attacks through Zero-shot Image Purification (ZIP). Our framework can be applied to poisoned models without requiring internal information about the model or any prior knowledge of the clean/poisoned samples. Our defense framework involves two steps. First, we apply a linear transformation (e.g., blurring) on the poisoned image to destroy the backdoor pattern. Then, we use a pre-trained diffusion model to recover the missing semantic information removed by the transformation. In particular, we design a new reverse process by using the transformed image to guide the generation of high-fidelity purified images, which works in zero-shot settings. We evaluate our ZIP framework on multiple datasets with different types of attacks. Experimental results demonstrate the superiority of our ZIP framework compared to state-of-the-art backdoor defense baselines. We believe that our results will provide valuable insights for future defense methods for black-box models. Our code is available at https://***/sycny/ZIP.

关键词：

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：