检索结果-内蒙古大学图书馆

Enhancing Offensive Language Detection with Data Augmentation and Knowledge Distillation

Research 2024年第2期6卷 29-40页

作者： Jiawen Deng Zhuang Chen Hao Sun Zhexin Zhang Jincenzi Wu Satoshi Nakagawa Fuji Ren Minlie Huang The CoAl group DCSTInstitute for Artificial InteligenceState Key Lab of Intelligent Technology and SystemsBeijing National Research Center for Information Science and TechnologyTsinghua UniversityBeijing 100084China Graduate School of Information Science&Technology The University of Tokyo.Tokyo 1138654Japan School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengduSichuan611731China

Offensive language detection has received important attention and plays a crucial role in promoting healthy communication on social platforms,as well as promoting the safe deployment of large language *** data is the basis for developing detectors;however,the available offense-related dataset in Chinese is severely limited in terms of data scale and coverage when compared to English *** significantly affects the accuracy of Chinese offensive language detectors in practical applications,especially when dealing with hard cases or out-of-domain *** alleviate the limitations posed by available datasets,we introduce AugCOLD(Augmented Chinese Offensive Language Dataset),a large-scale unsupervised dataset containing 1 million samples gathered by data crawling and model ***,we employ a multiteacher distillation framework to enhance detection performance with unsupervised *** is,we build multiple teachers with publicly accessible datasets and use them to assign soft labels to *** soft labels serve as a bridge for knowledge to be distilled from both AugCOLD and multiteacher to the student network,i.e.,the final offensive *** conduct experiments on multiple public test sets and our well-designed hard tests,demonstrating that our proposal can effectively improve the generalization and robustness of the offensive language detector.

关键词： COLD dealing limitations

来源：评论

学校读者我要写书评

暂无评论

Color liquid crystal grating based color holographic 3D display system with large viewing angle

引用

Light(Science & Applications) 2024年第1期13卷 103-112页

作者： Di Wang Yi-Long Li Fan Chu Nan-Nan Li Zhao-Song Li Sin-Doo Lee Zhong-Quan Nie Chao Liu Qiong-Hua Wang School of Instrumentation and Optoelectronic Engineering Beihang UniversityBeijing 100191China State Key Laboratory of Virtual Reality Technology and Systems Beihang UniversityBeijing 100191China Display Technology Research Center Seoul National UniversityGwanak-ro 1Gwanak-guSeoul 08826Republic of Korea Key Lab of Advanced Transducers and Intelligent Control System Ministry of EducationTaiyuan University of TechnologyTaiyuan 030024China

Holographic 3D display is highly desirable for numerous applications ranging from medical treatments to military ***,it is challenging to simultaneously achieve large viewing angle and high-fidelity color reconstruction due to the intractable constraints of existing ***,we conceptually propose and experimentally demonstrate a simple and feasible pathway of using a well-designed color liquid crystal grating to overcome the inevitable chromatic aberration and enlarge the holographic viewing angle,thus enabling large-viewing-angle and color holographic 3D *** use of color liquid crystal grating allows performing secondary diffraction modulation on red,green and blue reproduced images simultaneously and extending the viewing angle in the holographic 3D display *** principle,a chromatic aberration-free hologram generation mechanism in combination with the color liquid crystal grating is proposed to pave the way for on such a superior holographic 3D *** proposed system shows a color viewing angle of~50.12°,which is about 7 times that of the traditional system with a single spatial light *** work presents a paradigm for achieving desirable holographic 3D display,and is expected to provide a new way for the wide application of holographic display.

关键词： holographic angle crystal

来源：评论

学校读者我要写书评

暂无评论

Distilling Siamese Trackers with Attention Mask 41

Distilling Siamese Trackers with Attention Mask

引用

第41届中国控制会议

作者： Han Sun Yongqiang Bai Wenbo Zhang State Key Lab of Intelligent Control and Decision of Complex Systems and School of Automation Beijing Institute of Technology

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

In recent years,the introduction of Siamese network has brought new vitality to the object tracking ***,high-performance Siamese trackers cannot run at a real-time speed on mobile devices due to their complex and huge *** distillation is a common and effective model compression method,but it is difficult to be applied to the challenging task like object *** find out the fundamental cause is that the imbalance between the foreground and background in the object tracking task,which aggravates the problem of insufficient feature extraction ability of small ***,we propose the attention mask distillation(AMD) to help the student tracker focus on the foreground area faster and more *** attention mask can be easily obtained from the feature maps and brings fine-granularity to the traditional binary *** experimental results on OTB100 and VOT2018 show that our method enables the student tracker perform as well as the teacher *** the same time,it's able to run on the CPU at a hyper-real-time of 66 fps and achieves nearly 9 times model compression *** low computational and storage costs make it possible to deploy high-performance trackers on resource-constrained platforms.

关键词： Object Tracking Siamese Network Knowledge Distillation Attention Mask

来源：评论

学校读者我要写书评

暂无评论

Learning to Solve Multi-AGV Scheduling Problem with Pod Repositioning Optimization in RMFS 25

Learning to Solve Multi-AGV Scheduling Problem with Pod Repo...

引用

25th IEEE International Conference on Industrial technology, ICIT 2024

作者： Zhou, Xuan Shi, Xiang Chu, Wenqing Jiang, Jingchen Zhang, Lele Deng, Fang School Of Automation Beijing Institute Of Technology State Key Lab Of Autonomous Intelligent Unmanned Systems Beijing100081 China Beijing Institute Of Technology Chongqing Innovation Center Chongqing401120 China

ISBN: (纸本)9798350340266

This paper studies a novel non-deterministic polynomial-hard (NP-hard) warehouse optimization problem, the multi-automated guided vehicle (multi-AGV) scheduling problem with pod repositioning optimization (MAS-PR), in a robotic mobile fulfillment system (RMFS). A major challenge in MAS-PR is that the storage assignment of pods and the task allocation decision of AGVs are closely coupled. To address this problem, an end-to-end Problem-Specific Mask-based Deep Reinforcement Learning (PSMDRL) method is proposed in this paper. In PSMDRL, we transform the MAS-PR into a Markov Decision Process (MDP) model, and then use an attention-based network model to learn the efficient allocation and retrieval policy. A mask strategy with problem-specific knowledge and a model structure with decoder embedding are designed to further improve the efficiency and effectiveness. Experimental results demonstrate that the proposed PSMDRL is more effective and efficient than the representative algorithms in solving the MAS-PR problem. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Parameter Estimation and Tracking Control of MIMO Linear systems Without Prior Knowledge of Control Signs and Parameter Bounds

引用

IEEE Transactions on Automatic Control 2024年第6期70卷 3695-3710页

作者： Xu, Yuchun Zhang, Yanjun Zhang, Ji-Feng Chinese Academy of Sciences Key Laboratory of Systems and Control Institute of Systems Science Academy of Mathematics and Systems Science Beijing100190 China Beijing Institute of Technology School of Automation Beijing100081 China Beijing Institute of Technology State Key Lab of Automation Intelligent Unmanned Systems Beijing100081 China University of Chinese Academy of Sciences School of Mathematics Sciences Beijing100149 China Zhongyuan University of Technology School of Automation and Electrical Engineering Zhengzhou450007 China Chinese Academy of Sciences Academy of Mathematics and Systems Science Beijing100190 China

Dealing with the uncertain high-frequency gain matrix, denoted as Kp, is a fundamental problem in multivariable adaptive control systems. In this paper, we propose a new solution for parameter estimation and adaptive control for a general class of multi-input multi-output discrete-time linear time-invariant systems. The proposed scheme does not require any prior knowledge of the sign or bound information of Kp, and thus, significantly relaxes the design conditions in traditional multivariable adaptive control systems. Compared with the commonly used Nussbaum gain or multi-model techniques for addressing the unknown signs of Kp, the proposed scheme does not rely on any additional design conditions or any switching mechanism, while still ensuring closed-loop stability and asymptotic output tracking. Specifically, an output feedback adaptive control law is developed based on a matrix decomposition technique, which leads to derivation of a modified estimation error model. Subsequently, a gradient-based parameter update law is formulated only relying on the non-zero condition of the leading principle minors of Kp. Through designing gain functions and stable filters, the controller is always non-singular and does not involve any causal contradiction problem. Simulation study showcases the design process and demonstrates the effectiveness of the proposed scheme. © 1963-2012 IEEE.

关键词： MIMO systems

来源：评论

学校读者我要写书评

暂无评论

Traj-LLM: A New Exploration for Empowering Trajectory Prediction With Pre-Trained Large Language Models

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on intelligent Vehicles 2024年 1-14页

作者： Lan, Zhengxing Liu, Lingshan Fan, Bo Lv, Yisheng Ren, Yilong Cui, Zhiyong School of Transportation Science and Engineering Beihang University Beijing China Beijing Key Laboratory of Traffic Engineering College of Metropolitan Transportation Beijing University of Technology Beijing China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Transportation Science and Engineering State Key Lab of Intelligent Transportation System Beihang University Beijing China

Predicting the future trajectories of dynamic traffic actors is a cornerstone task in autonomous driving. Though existing notable efforts have resulted in impressive performance improvements, a gap persists in scene cognitive and understanding of complex traffic semantics. This paper proposes Traj-LLM, the first to investigate the potential of using pre-trained Large Language Models (LLMs) without explicit prompt engineering to generate future motions from vehicular past trajectories and traffic scene semantics. Traj-LLM starts with sparse context joint encoding to dissect the agent and scene features into a form that LLMs understand. On this basis, we creatively explore LLMs' strong understanding capability to capture a spectrum of high-level scene knowledge and interactive information. To emulate the human-like lane focus cognitive function and enhance Traj-LLM's scene comprehension, we introduce lane-aware probabilistic learning powered by the Mamba module. Finally, a multi-modal Laplace decoder is designed to achieve scene-compliant predictions. Extensive experiments manifest that Traj-LLM, fueled by prior knowledge and understanding prowess of LLMs, together with lane-aware probability learning, transcends the state-of-the-art methods across most evaluation metrics. Moreover, the few-shot analysis serves to substantiate Traj-LLM's performance, as even with merely 50% of the dataset, it surpasses the majority of benchmarks relying on complete data utilization. This study explores endowing the trajectory prediction task with advanced capabilities inherent in LLMs, furnishing a more universal and adaptable solution for forecasting agent movements in a new way. IEEE

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Instruction Pre-Training: Language Models are Supervised Mul...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Cheng, Daixuan Gu, Yuxian Huang, Shaohan Bi, Junyu Huang, Minlie Wei, Furu Microsoft Research United States The CoAI Group DCST Institute for Artificial Intelligence State Key Lab of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Tsinghua University China

ISBN: (纸本)9798891761643

Unsupervised multitask pre-training has been the critical method behind the recent success of language models (LMs). However, supervised multitask learning still holds significant promise, as scaling it in the post-training stage trends towards better generalization. In this paper, we explore supervised multitask pretraining by proposing Instruction Pre-Training, a framework that scalably augments massive raw corpora with instruction-response pairs to pre-train LMs. The instruction-response pairs are generated by an efficient instruction synthesizer built on open-source models. In our experiments, we synthesize 200M instruction-response pairs covering 40+ task categories to verify the effectiveness of Instruction Pre-Training. In pre-training from scratch, Instruction Pre-Training not only consistently enhances pre-trained base models but also benefits more from further instruction tuning. In continual pre-training, Instruction Pre-Training enables Llama3-8B to be comparable to or even outperform Llama3-70B. Our model, code, and data are available at https://***/microsoft/LMOps. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Superscattering of water waves

引用

National Science Review 2023年第7期10卷 91-97页

作者： Zijian Qin Chao Qian Lian Shen Xiaoping Wang Ido Kaminer Hongsheng Chen Huaping Wang Key Laboratory of Ocean Observation-Imaging Testbed of Zhejiang Province Institute of Marine Electronics Engineering Ocean College Zhejiang University ZJU-Hangzhou Global Science and Technology Innovation Center Key Lab.of Advanced Micro/Nano Electronic Devices & Smart Systems of Zhejiang Zhejiang University ZJU-UIUC Institute Interdisciplinary Center for Quantum Information State Key Laboratory of Modern Optical Instrumentation Zhejiang University Jinhua Institute of Zhejiang University Zhejiang University Department of Electrical Engineering Technion-Israel Institute of Technology

Inspired by the concept of superscattering in optics,we for the first time theoretically predict and experimentally demonstrate the superscattering phenomenon in water *** subwavelength superscatterer is constructed by multi-layered concentric cylinders with an inhomogeneous depth *** superscatterer breaks the long-held single-channel scattering limit by several times and thus significantly enhances the total scattering *** underlying mechanism originates from the near degeneracy of the resonances of multiple *** fabricate the superscatterer prototype and experimentally measure the near-field patterns,which are consistent with theoretical prediction and numerical *** study opens a new avenue to strengthen water-wave scattering and deepen the understanding in water waves,which can be useful for ocean energy harvesting and harbor protection.

关键词： superscattering water wave metamaterials

来源：评论

学校读者我要写书评

暂无评论

Recent Developments in Metallic Degradable Micromotors for Biomedical and Environmental Remediation Applications

引用

Nano-Micro Letters 2024年第3期16卷 1-35页

作者： Sourav Dutta Seungmin Noh Roger Sanchis Gual Xiangzhong Chen Salvador Pané Bradley J.Nelson Hongsoo Choi Department of Robotics and Mechatronics Engineering Daegu Gyeongbuk Institute of Science and Technology(DGIST)Daegu 42988Republic of Korea DGIST‑ETH Microrobotics Research Center DGISTDaegu 42988Republic of Korea Multi‑Scale Robotics Lab Institute of Robotics and Intelligent SystemsETH Zurich8092 ZurichSwitzerland Institute of Optoelectronics State Key Laboratory of Photovoltaic Science and TechnologyShanghai Frontiers Science Research Base of Intelligent Optoelectronics and PerceptionFudan UniversityShanghai 200433People’s Republic of China

Synthetic micromotor has gained substantial attention in biomedicine and environmental ***-based degradable micromotor composed of magnesium(Mg),zinc(Zn),and iron(Fe)have promise due to their nontoxic fuel-free propulsion,favorable biocompatibility,and safe excretion of degradation products Recent advances in degradable metallic micromotor have shown their fast movement in complex biological media,efficient cargo delivery and favorable biocompatibility.A noteworthy number of degradable metal-based micromotors employ bubble propulsion,utilizing water as fuel to generate hydrogen *** novel feature has projected degradable metallic micromotors for active in vivo drug delivery *** addition,understanding the degradation mechanism of these micromotors is also a key parameter for their design and *** propulsion efficiency and life span govern the overall performance of a degradable metallic *** we review the design and recent advancements of metallic degradable ***,we describe the controlled degradation,efficient in vivo drug delivery,and built-in acid neutralization capabilities of degradable micromotors with versatile biomedical ***,we discuss micromotors’efficacy in detecting and destroying environmental ***,we address the limitations and future research directions of degradable metallic micromotors.

关键词： Magnesium Zinc Iron Biodegradable microrobot Biomedical Environmental

来源：评论

学校读者我要写书评

暂无评论

Learning to Grasp through World Models with Backward Prediction

Learning to Grasp through World Models with Backward Predict...

引用

第43届中国控制会议

作者： Shuze Wang Yetian Yuan Yunpeng Mei Jian Sun Gang Wang State Key Lab of Autonomous Intelligent Unmanned Systems Beijing Institute of Technology

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

In the era of the fourth industrial revolution, the paradigm of intelligent manufacturing has ascended as a pivotal driver for future advancements. The combination of intelligent manufacturing with reinforcement learning offers significant benefits and opportunities. However, applying reinforcement learning in practical settings faces challenges due to sampling ***, model-based reinforcement learning algorithms have demonstrated higher sample efficiency. In this research, we introduce an enhanced version of the Dreamer algorithm and devise a system dedicated to mastering the tasks of picking and placing through the limited data collected. The improved algorithm addresses the issue of low sample efficiency and proves to be exceptionally effective in pixel-based tasks associated with high-dimensional state spaces. To corroborate the efficacy of the proposed algorithm, we evaluate the algorithm on a simulation environment of a production line for robotic arm operations developed in Coppeliasim. The merits of the augmented algorithm are further corroborated through its performance in complex DeepMind control tasks.

关键词： World Model Reinforcement Learning Robot Arm

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：