检索结果-内蒙古大学图书馆

End-to-End Streaming Customizable keyword Spotting Based on Text-Adaptive Neural Search 1

18th National Conference on Man-Machine Speech Communication, NCMMSC 2023

作者： Yang, Baochen Guo, Jiaqi Li, Haoyu Xi, Yu Zhuo, Qing Yu, Kai X-LANCE Lab Department of Computer Science and Engineering MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China State Key Laboratory of Media Convergence Production Technology and Systems Beijing China AISpeech Ltd. Suzhou China Department of Automation Tsinghua University Beijing China

ISBN: (数字)9789819706013

ISBN: (纸本)9789819706006

Streaming keyword spotting (KWS) is an important technique for voice assistant wake-up. While KWS with a preset fixed keyword has been well studied, test-time customizable keyword spotting in streaming mode remains a great challenge due to the lack of pre-collected keyword-specific training data and the requirement of streaming detection output. In this paper, we propose a novel end-to-end text-adaptive neural search architecture with a multi-label trigger mechanism to allow any pre-trained ASR acoustic model to be effectively used for fast streaming customizable keyword spotting. Evaluation results on various datasets show that our approach significantly outperforms both traditional post-processing baseline and the neural search baseline, meanwhile achieving a 44x search speedup compared to the traditional post-processing method. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Customizable keyword keyword spotting Streaming Text-adaptive search

来源：评论

学校读者我要写书评

暂无评论

Beyond Instruction Following: Evaluating Inferential Rule Following of Large Language Models

arXiv

引用

arXiv 2024年

作者： Sun, Wangtao Zhang, Chenxiang Zhang, Xueyou Yu, Xuanqing Huang, Ziyang Xu, Haotian Chen, Pei He, Shizhu Zhao, Jun Liu, Kang The Laboratory of Cognition and Decision Intelligence for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China CAS Engineering Laboratory for Intelligent Industrial Vision Institute of Automation Chinese Academy of Sciences Beijing China Department of Computer Science and Engineering Texas A&M University United States Shanghai Artificial Intelligence Laboratory China Xiaohongshu Inc China AI Lab AIGility Cloud Innovation Beijing China

Although Large Language Models (LLMs) have demonstrated strong instruction-following ability, they are further supposed to be controlled and guided by inferential rules in real-world scenarios to be safe, accurate, and intelligent. This demands the possession of inferential rule-following capability of LLMs. However, no prior work has made a clear evaluation of the inferential rule-following capability of LLMs. Previous studies that try to evaluate the inferential rule-following capability of LLMs fail to distinguish the inferential rule-following scenarios from the instruction-following scenarios. Therefore, this paper first clarifies the concept of inferential rule-following and proposes a comprehensive benchmark, RuleBench, to evaluate a diversified range of inferential rule-following abilities. Our experimental results on a variety of LLMs show that they are still limited in following rules. Our analysis based on the evaluation results provides insights into the improvements for LLMs toward a better inferential rule-following intelligent agent. We further propose Inferential Rule-Following Tuning (IRFT). The experimental results show that through IRFT, LLMs can learn abstract inferential rule-following abilities from purely synthetic data and then generalize to RuleBench. The data and code can be found at: https://***/r/llm-rulefollowing-B3E3/ Copyright © 2024, The Authors. All rights reserved.

关键词： Intelligent agents

来源：评论

学校读者我要写书评

暂无评论

CMFN: Cross-Modal Fusion Network for Irregular Scene Text Recognition

arXiv

引用

arXiv 2024年

作者： Zheng, Jinzhi Ji, Ruyi Zhang, Libo Wu, Yanjun Zhao, Chen Intelligent Software Research Center Institute of Software Chinese Academy of Sciences Beijing100190 China The Laboratory of Cognition and Decision Intelligence for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing China State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences Beijing100190 China

Scene text recognition, as a cross-modal task involving vision and text, is an important research topic in computer vision. Most existing methods use language models to extract semantic information for optimizing visual recognition. However, the guidance of visual cues is ignored in the process of semantic mining, which limits the performance of the algorithm in recognizing irregular scene text. To tackle this issue, we propose a novel cross-modal fusion network (CMFN) for irregular scene text recognition, which incorporates visual cues into the semantic mining process. Specifically, CMFN consists of a position self-enhanced encoder, a visual recognition branch and an iterative semantic recognition branch. The position self-enhanced encoder provides character sequence position encoding for both the visual recognition branch and the iterative semantic recognition branch. The visual recognition branch carries out visual recognition based on the visual features extracted by CNN and the position encoding information provided by the position self-enhanced encoder. The iterative semantic recognition branch, which consists of a language recognition module and a cross-modal fusion gate, simulates the way that human recognizes scene text and integrates cross-modal visual cues for text recognition. The experiments demonstrate that the proposed CMFN algorithm achieves comparable performance to stateof- the-art algorithms, indicating its effectiveness. © 2024, CC BY.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Cooperative Hierarchical Coverage Control of Multi-Agent systems with Non-holonomic Constraints in Poriferous Environments

引用

IEEE Transactions on Control of Network systems 2025年

作者： Zou, Jiayu Zhang, Hai-Tao Zhai, Chao Xing, Ning Ma, Yong Liu, Xingjian Huazhong University of Science and Technology School of Artificial Intelligence and Automation The MOE Engineering Research Center of Autonomous Intelligent Unmanned Systems The State Key Laboratory of Digital Manufacturing Equipment and Technology Wuhan430074 China China University of Geosciences School of Automation China The MOE Engineering Research Center of Intelligent Technology for Geo-Exploration Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Academician Workstation of China COSCO Shipping Corporation Ltd. Shanghai200120 China Wuhan University of Technology Sanya Science and Education Innovation Park Sanya572000 China Wuhan University of Technology Chongqing Research Institute Chongqing401120 China

It has long posed a challenging task to optimally deploy multi-agent systems (MASs) to cooperatively coverage poriferous environments in real cooperative detection applications. In response to this challenge, this paper proposes a hierarchical coverage control (HCC) protocol for MASs to perform sector-based coverage operations. First, a distributed Voronoi partition-based sweep-and-assign protocol, combined with a sectorial partition method, is developed, enabling the segmentation of a whole poriferous region into multiple sectorial sub-subregions. Following this procedure, an extreme search scheme is developed to determine the optimal amount of agents for each subregion. After this allocation, a distributed controller is proposed to deploy the agents, considering non-holonomic constraints, into designated niche positions. Additionally, sufficient conditions are derived to guarantee the asymptotical stability of the present HCC. The analytical challenge of the present study stems from the non-convex characteristics inherent in the loss function induced by porous environments. This non-convexity renders the task of globally optimizing the assignment of MASs intractable. Finally, numerical simulations are conducted to validate the effectiveness of the present coverage control approach. © 2014 IEEE.

关键词： Nonlinear control systems

来源：评论

学校读者我要写书评

暂无评论

AIRIS:Artificial intelligence Enhanced Signal Processing in Reconfigurable Intelligent Surface Communications

引用

China Communications 2021年第7期18卷 158-171页

作者： Shun Zhang Muye Li Mengnan Jian Yajun Zhao Feifei Gao State Key Laboratory of Integrated Services Networks Xidian UniversityXi’an 710071China Algorithm Department of Wireless Product R&D Institute ZTE CorporationBeijing 100029China Institute for Artificial Intelligence Tsinghua University(THUAI)State Key Lab of Intelligent Technologies and SystemsTsinghua UniversityBeijing National Research Center for Information Science and Technology(BNRist)Department of AutomationTsinghua University Beijing100084China

Reconfigurable intelligent surface(RIS)is an emerging meta-surface that can provide additional communications links through reflecting the signals,and has been recognized as a strong candidate of 6G mobile communications ***,it has been recently admitted that implementing artificial intelligence(AI)into RIS communications will extensively benefit the reconfiguration capacity and enhance the robustness to complicated transmission *** the conventional model-driven approaches,AI can also deal with the existing signal processing problems in a data-driven manner via digging the inherent characteristic from the real ***,AI is particularly suitable for the signal processing problems over RIS networks under unideal scenarios like modeling mismatching,insufficient resource,hardware impairment,as well as dynamical *** one of the earliest survey papers,we will introduce the merging of AI and RIS,called AIRIS,over various signal processing topics,including environmental sensing,channel acquisition,beamforming design,and resource scheduling,*** will also discuss the challenges of AIRIS and present some interesting future directions.

关键词： reconfigurable intelligent surface artifi-cial intelligence deep learning deep reinforcement learning signal processing

来源：评论

学校读者我要写书评

暂无评论

Asynchronous $\mathcal{L}_{1}$-gain Filtering of Nonlinear Positive Semi-Markov Jump systems with Time-varying Delays

Asynchronous $\mathcal{L}_{1}$-gain Filtering of Nonlinear P...

引用

Chinese Control Conference (CCC)

作者： Chao Ma Hang Fu Yidao Ji Wei Wu School of Mechanical Engineering University of Science and Technology Beijing China State Key Lab of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences China

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

In this paper, the $\mathcal{L}_{1}$ -gain based filtering problem for nonlinear positive semi-Markov jump systems is investigated by proposing a novel asynchronous design approach. More precisely, the mode-dependent filters are designed in terms of practical observed modes instead of true system modes, such that less conservatism can be achieved. In addition, the effect of time-varying delays is taken into account for more robustness and applicability. By selecting suitable stochastic Lyapunov-Krasovskii functions and applying the linear programming method, sufficient conditions are established to fulfill the desired $\mathcal{L}_{1}$ -gain performance. Eventually, the illustrative simulation is performed to verify the effectiveness of our developed control scheme.

关键词： Maximum likelihood detection Sufficient conditions Filtering Hidden Markov models Stochastic processes Nonlinear filters Linear programming

来源：评论

学校读者我要写书评

暂无评论

Locomotion control of a hybrid propulsion biomimetic underwater vehicle via deep reinforcement learning

Locomotion control of a hybrid propulsion biomimetic underwa...

引用

2021 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2021

作者： Zhang, Tiandong Wang, Rui Wang, Yu Wang, Shuo Institute of Automation Chinese Academy of Sciences State Key Laboratory of Management and Control for Complex Systems Beijing100190 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China

ISBN: (纸本)9781665436786

This paper presents a novel deep reinforcement learning (DRL) method to solve the locomotion control problem of the biomimetic underwater vehicle (BUV) with hybrid propulsion, in order to meet the challenge of intractable multi-fins coordination and the complex hydrodynamic model. The system overview of the BUV, named RoboDact, with two flexible long fins and a double-joint fishtail as hybrid propulsion, is introduced. After that, the locomotion control problem is modeled as a Markov decision process (MDP) to be solved. Therefore, the locomotion control method based on soft actor-critic (SAC, a novel DRL algorithm) is proposed. The simulation environment is established based on the kinetic model for interaction. Finally, the feasibility and effectiveness of the proposed control method is demonstrated after extensive simulations. It will provide rich insights into the coordination control of biomimetic underwater vehicles. © 2021 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Remote Operation with Haptic Force and Virtual Proxy for an Underwater Vehicle-Manipulator System 10

Remote Operation with Haptic Force and Virtual Proxy for an ...

引用

10th IEEE Data Driven Control and Learning systems Conference, DDCLS 2021

作者： Ma, Jin Wang, Yu Wang, Rui Wang, Shuo University of Chinese Academy of Sciences Beijing100049 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Cas Center for Excellence in Brain Science and Intelligence Technology Shanghai200031 China

ISBN: (纸本)9781665424233

This paper aims to investigate a smooth teleoperation method for the underwater vehicle-manipulator system. First, a coordinated mapping control method for the vehicle is presented. The haptic force is considered to help assist the operation. Then, two mapping modes are used to teleoperate the manipulator: when the end-effector needs to move in a large area, two virtual points and a spring-damping system are implemented to filter the operator's hand jitter and limit the manipulator's speed;when the end-effector needs to move in a small area, a position increment control method with a small proportional coefficient is used to improve the precision. Finally, the simulation demonstrates the effectiveness of the proposed teleoperation method. © 2021 IEEE.

关键词： Manipulators

来源：评论

学校读者我要写书评

暂无评论

RAG-based Crowdsourcing Task Decomposition via Masked Contrastive Learning with Prompts

arXiv

引用

arXiv 2024年

作者： Yang, Jing Wang, Xiao Zhao, Yu Liu, Yuhang Wang, Fei-Yue The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China The School of Artificial Intelligence Anhui University Hefei266114 China Qingdao Academy of Intelligent Industries Qingdao230031 China National Key Laboratory of Information Systems Engineering National University of Defense Technology China Macau University of Science and Technology 999078 China Beijing Engineering Research Center of Intelligent Systems and Technology Chinese Academy of Sciences Beijing100098 China State Key Laboratory for Management and Control of Complex Systems Chinese Academy of Sciences Beijing100190 China

Crowdsourcing is a critical technology in social manufacturing, which leverages an extensive and boundless reservoir of human resources to handle a wide array of complex tasks. The successful execution of these complex tasks relies on task decomposition (TD) and allocation, with the former being a prerequisite for the latter. Recently, pre-trained language models (PLMs)based methods have garnered significant attention. However, they are constrained to handling straightforward common-sense tasks due to their inherent restrictions involving limited and difficult-to-update knowledge as well as the presence of "hallucinations". To address these issues, we propose a retrieval-augmented generation-based crowdsourcing framework that reimagines TD as event detection from the perspective of natural language understanding. However, the existing detection methods fail to distinguish differences between event types and always depend on heuristic rules and external semantic analyzing tools. Therefore, we present a Prompt-Based Contrastive learning framework for TD (PBCT), which incorporates a prompt-based trigger detector to overcome dependence. Additionally, trigger-attentive sentinel and masked contrastive learning are introduced to provide varying attention to trigger and contextual features according to different event types. Experiment results demonstrate the competitiveness of our method in both supervised and zero-shot detection. A case study on printed circuit board manufacturing is showcased to validate its adaptability to unknown professional domains. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

GPR and SPSO-CG based gait pattern generation for subject-specific training

引用

science China(Information sciences) 2021年第8期64卷 244-246页

作者： Weiqun WANG Weiguo SHI Shixin REN Zeng-Guang HOU Xu LIANG Jiaxin WANG Liang PENG The State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of Sciences School of Artificial Intelligence University of Chinese Academy of Sciences CAS Center for Excellence in Brain Science and Intelligence Technology

Dear editor,Gait training has been proved effective for recovery of walking ability for nerve injury patients caused by stroke, spinal cord injury(SCI), traumatic brain injury(TBI), etc. The traditional gait training methods, usually carried out by hands, are laborious and time-consuming, and hence, the training intensity is difficult to be maintained and the rehabilitation results are not satisfactory.

关键词： GPR and SPSO-CG based gait pattern generation for subject-specific training

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：