检索结果-内蒙古大学图书馆

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Feng Wu Weiqi Wang Youyang Qu Shui Yu School of Computer Science University of Technology Sydney Sydney Australia Data61 Commonwealth Scientific and Industrial Research Organization Burwood Australia

ISBN: (数字)9798331506209

ISBN: (纸本)9798331506216

Note:This paper contains many malicious contents generated by LLMs. Jailbreak can cause large language models (LLMs) to violate moral and ethical guidelines, generating harmful text and images. However, the existing jailbreaks are high-cost (e.g., fine-tuning LLMs) and underperform when facing some specific jailbreak tasks (e.g., generating pornographic and bloody texts), which has too many limitations for reflecting the threat of jailbreak on LLMs’ security. To fill this gap, in this paper, we propose a black-box jailbreak with a higher attack success rate and lower cost, called Dictionary Jailbreak with Multi-round Induction (DJMI). First, in DJMI, we create a simple custom language dictionary based on specific jailbreak tasks to be performed and send it to the LLM. Then, we use the custom language from the dictionary to construct malicious prompts, instructing the LLM to respond in the custom language as much as possible. Generally, in the first round, the LLM will provide a neutral response (such as translating the malicious prompts) or refuse to answer. To counteract this, we need to emphasize that the prompts comply with ethical guidelines and repeat the prompts using the custom language. The process can be conducted with multiple rounds to guide LLM in outputting harmful content. During the attack process, the only attack cost is the transmission cost of the prompts through the official interactive interface, without any other costs of LLM fine-tuning, algorithm design and prompt generation. Thus, DJMI is a low-cost jailbreak method. We verified DJMI on multiple mainstream LLMs across various jailbreak tasks. Experiments show that compared to existing black-box jailbreaks, DJMI achieves a higher attack success rate (> 80% on average) across specific jailbreak tasks with various risk levels. Additionally, DJMI enables the LLM to provide highly detailed execution steps for harmful behaviors within a session (such as specifying proportions of ingredients, synthe

关键词： Drugs Ethics Waste materials Dictionaries Costs Translation Large language models Closed box Resists Guidelines

来源：评论

学校读者我要写书评

暂无评论

Neural Machine Translation with Contrastive Translation Memories

Neural Machine Translation with Contrastive Translation Memo...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Cheng, Xin Gao, Shen Liu, Lemao Zhao, Dongyan Yan, Rui Wangxuan Institute of Computer Technology Peking University China Center for Data Science Peking University China Tencent AI Lab School of Computer Science and Technology Shandong University China State Key Laboratory of Media Convergence Production Technology and Systems China China Gaoling School of Artificial Intelligence Renmin University of China China

Retrieval-augmented Neural Machine Translation models have been successful in many translation scenarios. Different from previous works that make use of mutually similar but redundant translation memories (TMs), we propose a new retrieval-augmented NMT to model contrastively retrieved translation memories that are holistically similar to the source sentence while individually contrastive to each other providing maximal information gains in three phases. First, in TM retrieval phase, we adopt a contrastive retrieval algorithm to avoid redundancy and uninformativeness of similar translation pieces. Second, in memory encoding stage, given a set of TMs we propose a novel Hierarchical Group Attention module to gather both local context of each TM and global context of the whole TM set. Finally, in training phase, a Multi-TM contrastive learning objective is introduced to learn salient feature of each TM with respect to target sentence. Experimental results show that our framework obtains improvements over strong baselines on the benchmark datasets. © 2022 Association for Computational Linguistics.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Prediction of traditional Chinese medicine prescriptions based on multi-label resampling

Journal of Electronic Business & Digital Economics

引用

Journal of Electronic Business & Digital Economics 2023年第2期2卷 213-227页

作者： Xiaomei Jiang Shuo Wang Wenjian Liu Yun Yang National Pilot School of Software Yunnan University Kunming China Yunnan Key Laboratory of Software Engineering Yunnan University Kunming China School of Computer Science University of Birmingham Birmingham UK School of Data Science City University of Macau Macau China

Purpose Traditional Chinese medicine (TCM) prescriptions have always relied on the experience of TCM doctors, and machine learning(ML) provides a technical means for learning these experiences and intelligently assists in prescribing. However, in TCM prescription, there are the main (Jun) herb and the auxiliary (Chen, Zuo and Shi) herb collocations. In a prescription, the types of auxiliary herbs are often more than the main herb and the auxiliary herbs often appear in other prescriptions. This leads to different frequencies of different herbs in prescriptions, namely, imbalanced labels (herbs). As a result, the existing ML algorithms are biased, and it is difficult to predict the main herb with less frequency in the actual prediction and poor performance. In order to solve the impact of this problem, this paper proposes a framework for multi-label traditional Chinese medicine (ML-TCM) based on multi-label resampling. Design/methodology/approach In this work, a multi-label learning framework is proposed that adopts and compares the multi-label random resampling (MLROS), multi-label synthesized resampling (MLSMOTE) and multi-label synthesized resampling based on local label imbalance (MLSOL), three multi-label oversampling techniques to rebalance the TCM data. Findings The experimental results show that after resampling, the less frequent but important herbs can be predicted more accurately. The MLSOL method is shown to be the best with over 10% improvements on average because it balances the data by considering both features and labels when resampling. Originality/value The authors first systematically analyzed the label imbalance problem of different sampling methods in the field of TCM and provide a solution. And through the experimental results analysis, the authors proved the feasibility of this method, which can improve the performance by 10%−30% compared with the state-of-the-art methods.

关键词： TCM prescription Multi-label classification Label imbalance Resampling

来源：评论

学校读者我要写书评

暂无评论

Counterfactual Explanation Through Constraint Relaxation

Counterfactual Explanation Through Constraint Relaxation

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Sharmi Dev Gupta Barry O'Sullivan Luis Quesada SFI Centre for Research Training in Artificial Intelligence School of Computer Science & IT University College Cork Cork Ireland Insight Centre for Data Analytics

ISBN: (数字)9798331527235

ISBN: (纸本)9798331527242

Interactive constraint systems often suffer from infeasibility (no solution) due to conflicting user constraints. A common approach to recover feasibility is to eliminate the constraints that cause the conflicts in the system. This approach allows the system to provide an explanation as: “if the user is willing to drop some of their constraints, there exists a solution”. However, this form of explanation might not be very informative. A counterfactual explanation is a type of explanation that can provide a basis for the user to recover feasibility by helping them understand what changes can be applied to their existing constraints rather than removing them. We propose an iterative method based on conflict detection and maximal relaxations in over-constrained constraint satisfaction problems to help compute a counterfactual explanation. We have evaluated our approach using well known instances that occur in industrial applications and demonstrated the relevance of multi-point relaxations.

关键词： programming Iterative methods Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

GyroCopter: Differential Bearing Measuring Trajectory Planner for Tracking and Localizing Radio Frequency Sources

arXiv

引用

arXiv 2024年

作者： Chen, Fei Rezatofighi, S. Hamid Ranasinghe, Damith C. The School of Computer & Mathematical Sciences The University of Adelaide SA5005 Australia The Department of Data Science & AI Monash University VIC3800 Australia

Autonomous aerial vehicles can provide efficient and effective solutions for radio frequency (RF) source tracking and localizing problems with applications ranging from wildlife conservation to search and rescue operations. Existing lightweight, low-cost, bearing measurements-based methods with a single antenna-receiver sensor system configurations necessitate in situ rotations, leading to substantial measurement acquisition times restricting searchable areas and number of measurements. We propose a GyroCopter for the task. Our approach plans the trajectory of a multi-rotor unmanned aerial vehicle (UAV) whilst utilizing UAV flight dynamics to execute a constant gyration motion to derive "pseudo-bearing" measurements to track RF sources. The gyration-based pseudo-bearing approach: i) significantly reduces the limitations associated with in situ rotation bearing;while ii) capitalizing on the simplicity, affordability, and lightweight nature of signal strength measurement acquisition hardware to estimate bearings. This method distinguishes itself from other pseudo-bearing approaches by eliminating the need for additional hardware to maintain simplicity, lightweightness and cost-effectiveness. To validate our approach, we derived the optimal rotation speed and conducted extensive simulations and field missions with our GyroCopter to track and localize multiple RF sources. The results confirm the effectiveness of our method, highlighting its potential as a practical and rapid solution for RF source localization tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

ProMamba: Prompt-Mamba for polyp segmentation

arXiv

引用

arXiv 2024年

作者： Xie, Jianhao Liao, Ruofan Zhang, Ziang Yi, Sida Zhu, Yuesheng Luo, Guibo School of Electronic and Computer Engineering Peking University China Aberdeen Institute of Data Science and Artificial Intelligence South China Normal University China

Detecting polyps through colonoscopy is an important task in medical image segmentation, which provides significant assistance and reference value for clinical surgery. However, accurate segmentation of polyps is a challenging task due to two main reasons. Firstly, polyps exhibit various shapes and colors. Secondly, the boundaries between polyps and their normal surroundings are often unclear. Additionally, significant differences between different datasets lead to limited generalization capabilities of existing methods. To address these issues, we propose a segmentation model based on Prompt-Mamba, which incorporates the latest Vision-Mamba and prompt technologies. Compared to previous models trained on the same dataset, our model not only maintains high segmentation accuracy on the validation part of the same dataset but also demonstrates superior accuracy on unseen datasets, exhibiting excellent generalization capabilities. Notably, we are the first to apply the Vision-Mamba architecture to polyp segmentation and the first to utilize prompt technology in a polyp segmentation model. Our model efficiently accomplishes segmentation tasks, surpassing previous state-of-the-art methods by an average of 5% across six datasets. Furthermore, we have developed multiple versions of our model with scaled parameter counts, achieving better performance than previous models even with fewer parameters. Our code and trained weights will be released soon. Copyright © 2024, The Authors. All rights reserved.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

The Emergent Aging Model: Aging as an Emergent Property of Biological Systems

arXiv

引用

arXiv 2024年

作者： Qin, Hong Department of Computer Science and Engineering Department of Biology Geology and Environmental Sciences University of Tennessee at Chattanooga ChattanoogaTN United States Department of Computer Science School of Data Sciences Old Dominion University NorfolkVA United States

Based on the study of cellular aging using the single cell model organism of budding yeast and corroborated by other studies, we propose the Emergent Aging Model (EAM). EAM hypothesizes that aging is an emergent property of complex biological systems, exemplified by biological networks such as gene networks. An emergent property refers to traits that a system has at the system level but which its low-level components do not. EAM is based on a quantitative definition of aging using the mortality rate. A biological entity with a constant mortality rate is considered non-aging which is equivalent to a first-order chemical reaction. Aging can be quantitatively defined an increasing mortality rate over time, corresponding to an organism’s increasing chance of dying over time. EAM posits that biological aging can arise at the system level of an organism, even if the system is composed of only non-aging components. EAM is consistent with the observation that aging is largely stochastic, influenced by numerous genes and epigenetic factors, with no single gene or factor known as the bone fide cause of aging. A parsimonious version of EAM can predict the Gompertz model of biological aging, the Strehler-Mildvan correlation, and the trade-off between initial reproductive fitness (asexual reproductive fitness) and late-life survival. EAM has been applied to experimental results of the replicative lifespan of the budding yeast and can potentially offer new insights into the aging process of other biological species. © 2024, CC BY.

关键词： Population statistics

来源：评论

学校读者我要写书评

暂无评论

Unlocking Consumer Minds: Navigating Purchase Intentions through the Lens of Augmented Reality and the Technology Acceptance Model

Unlocking Consumer Minds: Navigating Purchase Intentions thr...

引用

Technology, Engineering, and Computing Applications (ICTECA), International Conference on

作者： Dian Martha Nurrul Amanah Nurul Aini Novinda Krisna Putri Adhe Lingga Dewi Jenia Nur Soelistyoningrum Digital Business Department BINUS Business School Undergraduate Program Bina Nusantara University Semarang Indonesia Computer Science Department School of Computer Science Bina Nusantara University Semarang Indonesia

As a relatively new technology, augmented reality can help business continuity. This technology can maximize various business activities, one of which is marketing. Unfortunately, this technology is still rarely used for business needs in Indonesia. Hence, this paper aims to investigate how the technology acceptance model affects augmented reality-based marketing applications and their impact on the intention to make a purchase. The research adopted a quantitative methodology, specifically employing survey techniques. The primary data gathered underwent analysis through Partial Least Squares Structural Equation Modeling (PLS-SEM). The process of data collection included the distribution of online questionnaires to individuals actively utilizing smartphones. These participants were subsequently requested to engage with one of the five selected augmented reality marketing applications: Shopee (Beauty Cam), Sephora (Virtual Artist), IKEA Place, and Dulux (Visualizer). To uphold result reliability, the study encompassed a total of 109 respondents. The study revealed that perceived usefulness, and perceived enjoyment has a positive influence on purchase intention. Besides that, perceived ease of use has a negative influence on purchase intention. Based on the results of this research, it is important for business to be able to capture profitable opportunities from adopting augmented reality technology. So that in the future they will be able to provide the best experience for consumers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Theoretical Insights for Diffusion Guidance: A Case Study for Gaussian Mixture Models

arXiv

引用

arXiv 2024年

作者： Wu, Yuchen Chen, Minshuo Li, Zihao Wang, Mengdi Wei, Yuting Department of Statistics and Data Science The Wharton School University of Pennsylvania United States Department of Electrical and Computer Engineering Princeton University United States

Diffusion models benefit from instillation of task-specific information into the score function to steer the sample generation towards desired properties. Such information is coined as guidance. For example, in text-to-image synthesis, text input is encoded as guidance to generate semantically aligned images. Proper guidance inputs are closely tied to the performance of diffusion models. A common observation is that strong guidance promotes a tight alignment to the task-specific information, while reducing the diversity of the generated samples. In this paper, we provide the first theoretical study towards understanding the influence of guidance on diffusion models in the context of Gaussian mixture models. Under mild conditions, we prove that incorporating diffusion guidance not only boosts classification confidence but also diminishes distribution diversity, leading to a reduction in the differential entropy of the output distribution. Our analysis covers the widely adopted sampling schemes including DDPM and DDIM, and leverages comparison inequalities for differential equations as well as the Fokker-Planck equation that characterizes the evolution of probability density function, which may be of independent theoretical interest. Copyright © 2024, The Authors. All rights reserved.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

A Low-cost Black-box Jailbreak Based on Custom Mapping Dictionary with Multi-round Induction 23

A Low-cost Black-box Jailbreak Based on Custom Mapping Dicti...

引用

23rd IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2024

作者： Wu, Feng Wang, Weiqi Qu, Youyang Yu, Shui University of Technology Sydney School of Computer Science Sydney Australia Data61 Commonwealth Scientific and Industrial Research Organization Burwood Australia

Note:This paper contains many malicious contents generated by LLMs. Jailbreak can cause large language models (LLMs) to violate moral and ethical guidelines, generating harmful text and images. However, the existing jailbreaks are high-cost (e.g., fine-tuning LLMs) and underperform when facing some specific jailbreak tasks (e.g., generating pornographic and bloody texts), which has too many limitations for reflecting the threat of jailbreak on LLMs' security. To fill this gap, in this paper, we propose a black-box jailbreak with a higher attack success rate and lower cost, called Dictionary Jailbreak with Multi-round Induction (DJMI). First, in DJMI, we create a simple custom language dictionary based on specific jailbreak tasks to be performed and send it to the LLM. Then, we use the custom language from the dictionary to construct malicious prompts, instructing the LLM to respond in the custom language as much as possible. Generally, in the first round, the LLM will provide a neutral response (such as translating the malicious prompts) or refuse to answer. To counteract this, we need to emphasize that the prompts comply with ethical guidelines and repeat the prompts using the custom language. The process can be conducted with multiple rounds to guide LLM in outputting harmful content. During the attack process, the only attack cost is the transmission cost of the prompts through the official interactive interface, without any other costs of LLM fine-tuning, algorithm design and prompt generation. Thus, DJMI is a low-cost jailbreak method. We verified DJMI on multiple mainstream LLMs across various jailbreak tasks. Experiments show that compared to existing black-box jailbreaks, DJMI achieves a higher attack success rate (> 80% on average) across specific jailbreak tasks with various risk levels. Additionally, DJMI enables the LLM to provide highly detailed execution steps for harmful behaviors within a session (such as specifying proportions of ingredients, synthe

关键词： black-box jailbreak jailbreak tasks LLMs prompts

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：