检索结果-内蒙古大学图书馆

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

作者： Wenkai Yang Xiaohan Bi Yankai Lin Sishuo Chen Jie Zhou Xu Sun Gaoling School of Artificial Intelligence Renmin University of China Beijing China Center for Data Science Peking University Pattern Recognition Center WeChat AI Tencent Inc. China National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University

ISBN: (纸本)9798331314385

Driven by the rapid development of Large Language Models (LLMs), LLM-based agents have been developed to handle various real-world applications, including finance, healthcare, and shopping, etc. It is crucial to ensure the reliability and security of LLM-based agents during applications. However, the safety issues of LLM-based agents are currently under-explored. In this work, we take the first step to investigate one of the typical safety threats, backdoor attack, to LLM-based agents. We first formulate a general framework of agent backdoor attacks, then we present a thorough analysis of different forms of agent backdoor attacks. Specifically, compared with traditional backdoor attacks on LLMs that are only able to manipulate the user inputs and model outputs, agent backdoor attacks exhibit more diverse and covert forms: (1) From the perspective of the final attacking outcomes, the agent backdoor attacker can not only choose to manipulate the final output distribution, but also introduce the malicious behavior in an intermediate reasoning step only, while keeping the final output correct. (2) Furthermore, the former category can be divided into two subcategories based on trigger locations, in which the backdoor trigger can either be hidden in the user query or appear in an intermediate observation returned by the external environment. We implement the above variations of agent backdoor attacks on two typical agent tasks including web shopping and tool utilization. Extensive experiments show that LLM-based agents suffer severely from backdoor attacks and such backdoor vulnerability cannot be easily mitigated by current textual backdoor defense algorithms. This indicates an urgent need for further research on the development of targeted defenses against backdoor attacks on LLM-based agents. Code and data are available at https://***/lancopku/agent-backdoor-attacks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

RECALL: A Benchmark for LLMs Robustness against External Counterfactual Knowledge

arXiv

引用

arXiv 2023年

作者： Liu, Yi Huang, Lianzhe Li, Shicheng Chen, Sishuo Zhou, Hao Meng, Fandong Zhou, Jie Sun, Xu National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China Pattern Recognition Center WeChat AI Tencent Inc. China Center for Data Science Peking University China

LLMs and AI chatbots have improved people\'s efficiency in various ***, the necessary knowledge for answering the question may be beyond themodels\' knowledge boundaries. To mitigate this issue, many researchers try tointroduce external knowledge, such as knowledge graphs and Internet contents,into LLMs for up-to-date information. However, the external information fromthe Internet may include counterfactual information that will confuse the modeland lead to an incorrect response. Thus there is a pressing need for LLMs topossess the ability to distinguish reliable information from externalknowledge. Therefore, to evaluate the ability of LLMs to discern thereliability of external knowledge, we create a benchmark from existingknowledge bases. Our benchmark consists of two tasks, Question Answering andText Generation, and for each task, we provide models with a context containingcounterfactual information. Evaluation results show that existing LLMs aresusceptible to interference from unreliable external knowledge withcounterfactual information, and simple intervention methods make limitedcontributions to the alleviation of this issue. © 2023, CC BY-NC-SA.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Zero-Shot Audio Captioning Using Soft and Hard Prompts

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language processing 2025年 33卷 2045-2058页

作者： Yiming Zhang Xuenan Xu Ruoyi Du Haohe Liu Yuan Dong Zheng-Hua Tan Wenwu Wang Zhanyu Ma Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Centre for Vision Speech and Signal Processing University of Surrey Guildford U.K. Department of Electronic Systems Aalborg University Aalborg Denmark

In traditional audio captioning methods, a model is usually trained in a fully supervised manner using a human-annotated dataset containing audio-text pairs and then evaluated on the test set from the same dataset. Such methods have two limitations. First, these methods are often data-hungry and require time-consuming and expensive human annotations to obtain audio-text pairs. Second, these models often suffer from performance degradation in cross-domain scenarios, i.e., when the input audio comes from a different domain than the training set, and this issue has received little attention. To address these issues, we propose a new zero-shot method for audio captioning. Our method is built on the contrastive language-audio pre-training (CLAP) model. During training, the model reconstructs the ground-truth caption using the CLAP text encoder. In the inference stage, the model generates text descriptions from the CLAP audio embeddings of given audio inputs. To enhance the ability of the model in transitioning from text-to-text generation to audio-to-text generation, we propose to use the mixed-augmentations-based soft prompt to learn more robust latent representations, leveraging instance replacement and embedding augmentation. Additionally, we introduce the retrieval-based acoustic-aware hard prompt to improve the cross-domain performance of the model by employing the domain-agnostic label information of sound events. Extensive experiments on AudioCaps and Clotho benchmarks show the effectiveness of our proposed method, which outperforms other zero-shot audio captioning approaches for in-domain scenarios and outperforms the compared methods for cross-domain scenarios, underscoring the generalization ability of our method.

关键词： Training Decoding Semantics Data models Acoustics Electronic mail Benchmark testing Transformers Robustness Perturbation methods

来源：评论

学校读者我要写书评

暂无评论

ENHANCING KERNEL FLEXIBILITY VIA LEARNING ASYMMETRIC LOCALLY-ADAPTIVE KERNELS

arXiv

引用

arXiv 2023年

作者： He, Fan He, Mingzhen Shi, Lei Huang, Xiaolin Suykens, Johan A.K. STADIUS Center for Dynamical Systems Signal Processing and Data Analytics KU Leuven Belgium Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University China Shanghai Key Laboratory for Contemporary Applied Mathematics School of Mathematical Sciences Fudan University China

The lack of sufficient flexibility is the key bottleneck of kernel-based learning that relies on manually designed, pre-given, and non-trainable kernels. To enhance kernel flexibility, this paper introduces the concept of Locally-Adaptive-Bandwidths (LAB) as trainable parameters to enhance the Radial Basis Function (RBF) kernel, giving rise to the LAB RBF kernel. The parameters in LAB RBF kernels are data-dependent, and its number can increase with the dataset, allowing for better adaptation to diverse data patterns and enhancing the flexibility of the learned function. This newfound flexibility also brings challenges, particularly with regards to asymmetry and the need for an efficient learning algorithm. To address these challenges, this paper for the first time establishes an asymmetric kernel ridge regression framework and introduces an iterative kernel learning algorithm. This novel approach not only reduces the demand for extensive support data but also significantly improves generalization by training bandwidths on the available training data. Experimental results on real datasets underscore the remarkable performance of the proposed algorithm, showcasing its superior capability in handling large-scale datasets compared to Nyström approximation-based algorithms. Moreover, it demonstrates a significant improvement in regression accuracy over existing kernel-based learning methods and even surpasses residual neural networks. Copyright © 2023, The Authors. All rights reserved.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents

arXiv

引用

arXiv 2024年

作者： Yang, Wenkai Bi, Xiaohan Lin, Yankai Chen, Sishuo Zhou, Jie Sun, Xu Gaoling School of Artificial Intelligence Renmin University of China Beijing China Center for Data Science Peking University China Pattern Recognition Center WeChat AI Tencent Inc. China National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China

Driven by the rapid development of Large Language Models (LLMs), LLM-based agents have been developed to handle various real-world applications, including finance, healthcare, and shopping, etc. It is crucial to ensure the reliability and security of LLM-based agents during applications. However, the safety issues of LLM-based agents are currently under-explored. In this work, we take the first step to investigate one of the typical safety threats, backdoor attack, to LLM-based agents. We first formulate a general framework of agent backdoor attacks, then we present a thorough analysis of different forms of agent backdoor attacks. Specifically, compared with traditional backdoor attacks on LLMs that are only able to manipulate the user inputs and model outputs, agent backdoor attacks exhibit more diverse and covert forms: (1) From the perspective of the final attacking outcomes, the agent backdoor attacker can not only choose to manipulate the final output distribution, but also introduce the malicious behavior in an intermediate reasoning step only, while keeping the final output correct. (2) Furthermore, the former category can be divided into two subcategories based on trigger locations, in which the backdoor trigger can either be hidden in the user query or appear in an intermediate observation returned by the external environment. We implement the above variations of agent backdoor attacks on two typical agent tasks including web shopping and tool utilization. Extensive experiments show that LLM-based agents suffer severely from backdoor attacks and such backdoor vulnerability cannot be easily mitigated by current textual backdoor defense algorithms. This indicates an urgent need for further research on the development of targeted defenses against backdoor attacks on LLM-based agents.3 Warning: This paper may contain biased content. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

TOWARDS CODABLE WATERMARKING FOR INJECTING MULTI-BITS INFORMATION TO LLMS

arXiv

引用

arXiv 2023年

作者： Wang, Lean Yang, Wenkai Chen, Deli Zhou, Hao Lin, Yankai Meng, Fandong Zhou, Jie Sun, Xu National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China Gaoling School of Artificial Intelligence Renmin University of China China Pattern Recognition Center WeChat AI Tencent Inc. China DeepSeek AI

As large language models (LLMs) generate texts with increasing fluency and realism, there is a growing need to identify the source of texts to prevent the abuse of LLMs. Text watermarking techniques have proven reliable in distinguishing whether a text is generated by LLMs by injecting hidden patterns. However, we argue that existing LLM watermarking methods are encoding-inefficient and cannot flexibly meet the diverse information encoding needs (such as encoding model version, generation time, user id, etc.). In this work, we conduct the first systematic study on the topic of Codable Text Watermarking for LLMs (CTWL) that allows text watermarks to carry multi-bit customizable information. First of all, we study the taxonomy of LLM watermarking technologies and give a mathematical formulation for CTWL. Additionally, we provide a comprehensive evaluation system for CTWL: (1) watermarking success rate, (2) robustness against various corruptions, (3) coding rate of payload information, (4) encoding and decoding efficiency, (5) impacts on the quality of the generated text. To meet the requirements of these non-Pareto-improving metrics, we follow the most prominent vocabulary partition-based watermarking direction, and devise an advanced CTWL method named Balance-Marking. The core idea of our method is to use a proxy language model to split the vocabulary into probability-balanced parts, thereby effectively maintaining the quality of the watermarked text. Extensive experimental results show that our method outperforms the baseline under comprehensive evaluation. Our code is available at https://***/lancopku/codable-watermarking-for-llm. © 2023, CC BY.

关键词： signal encoding

来源：评论

学校读者我要写书评

暂无评论

Riemannian Self-Attention Mechanism for SPD Networks

arXiv

引用

arXiv 2023年

作者： Wang, Rui Wu, Xiao-Jun Li, Hui Kittler, Josef School of Artificial Intelligence and Computer Science Jiangnan University Wuxi214122 China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University China Centre for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

Symmetric positive definite (SPD) matrix has been demonstrated to be an effective feature descriptor in many scientific areas, as it can encode spatiotemporal statistics of the data adequately on a curved Riemannian manifold, i.e., SPD manifold. Although there are many different ways to design network architectures for SPD matrix nonlinear learning, very few solutions explicitly mine the geometrical dependencies of features at different layers. Motivated by the great success of self-attention mechanism in capturing long-range relationships, an SPD manifold self-attention mechanism (SMSA) is proposed in this paper using some manifold-valued geometric operations, mainly the Riemannian metric, Riemannian mean, and Riemannian optimization. Then, an SMSA-based geometric learning module (SMSA-GLM) is designed for the sake of improving the discrimination of the generated deep structured representations. Extensive experimental results achieved on three benchmarking datasets show that our modification against the baseline network further alleviates the information degradation problem and leads to improved accuracy. Copyright © 2023, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Zero-Shot Audio Captioning Using Soft and Hard Prompts

arXiv

引用

arXiv 2024年

作者： Zhang, Yiming Xu, Xuenan Du, Ruoyi Liu, Haohe Dong, Yuan Tan, Zheng-Hua Wang, Wenwu Ma, Zhanyu The Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China The Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China The Department of Electronic Systems Aalborg University Aalborg9220 Denmark The Centre for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

In traditional audio captioning methods, a model is usually trained in a fully supervised manner using a human-annotated dataset containing audio-text pairs and then evaluated on the test sets from the same dataset. Such methods have two limitations. First, these methods are often data-hungry and require time-consuming and expensive human annotations to obtain audio-text pairs. Second, these models often suffer from performance degradation in cross-domain scenarios, i.e., when the input audio comes from a different domain than the training set, which, however, has received little attention. We propose an effective audio captioning method based on the contrastive language-audio pre-training (CLAP) model to address these issues. Our proposed method requires only textual data for training, enabling the model to generate text from the textual feature in the cross-modal semantic space. In the inference stage, the model generates the descriptive text for the given audio from the audio feature by leveraging the audio-text alignment from CLAP. We devise two strategies to mitigate the discrepancy between text and audio embeddings: a mixed-augmentation-based soft prompt and a retrieval-based acoustic-aware hard prompt. These approaches are designed to enhance the generalization performance of our proposed model, facilitating the model to generate captions more robustly and accurately. Extensive experiments on AudioCaps and Clotho benchmarks show the effectiveness of our proposed method, which outperforms other zero-shot audio captioning approaches for in-domain scenarios and outperforms the compared methods for cross-domain scenarios, underscoring the generalization ability of our method. © 2024, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Parameter estimation of photovoltaic modules using analytical and numerical/iterative approaches: A comparative study

引用

Materials Today: Proceedings 2022年 52卷 1-6页

作者： Souad Lidaighbi Mustapha Elyaqouti Khalid Assalaou Dris Ben Hmamou Driss Saadaoui Jihad H'roura Materials and Renewable Energy Laboratory Agadir Faculty of Sciences Ibn Zohr University BP 8106 80000 Agadir Morocco Laboratory of Electronics Signal Processing and Physical Modeling Faculty of Sciences of Agadir Ibn Zohr University BP 8106 80000 Agadir Morocco Laboratory of Images and Pattern Recognition - Intelligent and Communicating Systems Faculty of Sciences of Agadir Ibn Zohr University BP 8106 80000 Agadir Morocco

Modeling photovoltaic (PV) modules is fundamental for analyzing their efficiency and performance under different operating conditions. Generally, photovoltaic module modeling is based on a suitable equivalent circuit and a set of parameters representing the PV module's properties. Defining these parameters is not a trivial task, as they are usually not included in the technical documentation of the module. Therefore, the need to find suitable values for the model parameters becomes imperative. In this paper, we will test and compare two methods of extracting single-diode model (SDM) parameters: analytical and numerical, under standard test conditions (STC conditions). We examined the performance of six techniques: three analytical and three numerical. Except for Saleem technique, which tends to underestimate the I-V and P-V curves in the MPP zone, the results demonstrated that both approaches (analytical and iterative/numerical) are valid and accurate for predicting the I-V and P-V curves under STC conditions.

关键词： Photovoltaic module Single-diode model Parameters extraction Analytical methods Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Exploring Fusion Strategies for Accurate RGBT Visual Object Tracking

arXiv

引用

arXiv 2022年

作者： Tang, Zhangyong Xu, Tianyang Li, Hui Wu, Xiao-Jun Zhu, XueFeng Kittler, Josef Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence School of Artificial Intelligence and Computer Science Jiangnan University Wuxi214122 China The Center for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

We address the problem of multi-modal object tracking in video and explore various options of fusing the complementary information conveyed by the visible (RGB) and thermal infrared (TIR) modalities including pixel-level, feature-level and decision-level fusion. Specifically, different from the existing methods, paradigm of image fusion task is heeded for fusion at pixel level. Feature-level fusion is fulfilled by attention mechanism with channels excited optionally. Besides, at decision level, a novel fusion strategy is put forward since an effortless averaging configuration has shown the superiority. The effectiveness of the proposed decision-level fusion strategy owes to a number of innovative contributions, including a dynamic weighting of the RGB and TIR contributions and a linear template update operation. A variant of which produced the winning tracker at the Visual Object Tracking Challenge 2020 (VOT-RGBT2020). The concurrent exploration of innovative pixel- and feature-level fusion strategies highlights the advantages of the proposed decision-level fusion method. Extensive experimental results on three challenging datasets, i.e., GTOT, VOT-RGBT2019, and VOT-RGBT2020, demonstrate the effectiveness and robustness of the proposed method, compared to the state-of-the-art approaches. Code will be shared at https://***/Zhangyong-Tang/DFAT. © 2022, CC BY.

关键词： Pixels

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：