检索结果-南通市图书馆

Developing ChatGPT for biology and medicine: a complete review of biomedical question answering

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Biophysics Reports 2024年第3期10卷 152-171页

作者： Qing Li Lei Li Yu Li department of computer science and engineering the Chinese University of Hong Kong

ChatGPT explores a strategic blueprint of question answering(QA) to deliver medical diagnoses, treatment recommendations, and other healthcare support. This is achieved through the increasing incorporation of medical domain data via natural language processing(NLP) and multimodal paradigms. By transitioning the distribution of text, images, videos, and other modalities from the general domain to the medical domain, these techniques have accelerated the progress of medical domain question answering(MDQA). They bridge the gap between human natural language and sophisticated medical domain knowledge or expert-provided manual annotations, handling large-scale, diverse, unbalanced, or even unlabeled data analysis scenarios in medical contexts. Central to our focus is the utilization of language models and multimodal paradigms for medical question answering, aiming to guide the research community in selecting appropriate mechanisms for their specific medical research *** tasks such as unimodal-related question answering, reading comprehension, reasoning, diagnosis, relation extraction, probability modeling, and others, as well as multimodal-related tasks like vision question answering, image captioning, cross-modal retrieval, report summarization, and generation, are discussed in detail. Each section delves into the intricate specifics of the respective method under consideration. This paper highlights the structures and advancements of medical domain explorations against general domain methods, emphasizing their applications across different tasks and datasets. It also outlines current challenges and opportunities for future medical domain research,paving the way for continued innovation and application in this rapidly evolving field. This comprehensive review serves not only as an academic resource but also delineates the course for future probes and utilization in the field of medical question answering.

关键词： ChatGPT Medical question answering Nature language processing Multimodal paradigms Large language models

Cooperative Target Fencing of Multiple Vehicles for a General Target with Connectivity Preservation and Collision Avoidance

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Systems science & Complexity 2024年第1期37卷 136-151页

作者： PAN Zini CHEN Ben M department of Mechanical and Automation engineering The Chinese University of Hong KongHong KongChina

In this paper,the authors study the cooperative target-fencing problem for n-dimensional systems and a target with a general *** using the velocity of the vehicles,a position feedback control law is proposed to fence the general target into the convex hull formed by the ***,the dynamics of each vehicle is described by a double-integrator *** potential functions are designed to guarantee connectivity preservation of the communication network and collision avoidance among the *** proposed approach can deal with a target whose trajectory is any twice continuously differentiable function of *** effectiveness of the result is verified by a numerical example.

关键词： Multi-agent systems networked systems target fencing

Selective deposition of a MOF at the spikes of Au nanostars for SERS detection

在线全文

学校读者我要写书评

暂无评论

Nano Research 2024年第10期17卷 9166-9173页

作者： Yi Liu Ka Kit Chui Xinyue Xia Han Zhang Xiaolu Zhuo Jianfang Wang School of science and engineering The Chinese University of Hong KongShenzhen 518172China department of Physics The Chinese University of Hong KongShatinHong Kong 999077China School of Materials science and engineering Zhejiang Sci-Tech UniversityHangzhou 310000China

In the pursuit of advancing molecular sensing through surface-enhanced Raman spectroscopy(SERS),the combination of plasmonic nanoparticles and metal-organic frameworks(MOFs)has emerged as a highly effective approach to enhance the sensitivity and selectivity of SERS ***,most prior investigations have predominantly focused on MOF-coated plasmonic nanoparticles in core@shell or layer-by-layer configurations,leaving a notable knowledge gap in exploring alternative *** we present a facile method to construct a particle-on-mirror architecture by selectively coating a MOF,zeolitic imidazolate framework-8(ZIF-8),onto the tips of Au nanostars and subsequently depositing the resultant nanoparticles onto a Au *** design integrates the electric field enhancement at the sharp tips and nanogaps,along with the molecular enrichment function within the porous MOF immobilized at the tips and nanogaps,leading to a substantial boost in the SERS signal *** a unique SERS platform enables consistent and outstanding SERS performance for analytes of different *** work opens up a promising strategy for constructing multifunctional nanostructures for sensitive SERS detection in real-life scenarios.

关键词： electric field enhancement gold nanostars metal-organic frameworks molecular enrichment plasmon coupling surfaceenhanced Raman spectroscopy(SERS)substrates

Hybrid vibration isolation optimization of a flexible manipulator based on neural network agent model

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

chinese Journal of Aeronautics 2024年第5期37卷 482-505页

作者： Yongxin ZHANG Liang LI Dingguo ZHANG Wei-Hsin LIAO Xian GUO School of science Nanjing University of Science and TechnologyNanjing 210094China department of Mechanical and Automation engineering The Chinese University of Hong KongHong KongChina

Segmented Active Constrained Layer Damping(SACLD)is an intelligent vibration-damping structure,which could be applied to the sectors of aviation,aerospace,and transportation engineering to reduce the vibration of flexible ***,machine learning technology is widely used in the engineering field because of its efficient multi-objective *** dynamic simulation of a rotational segmental flexible manipulator system is presented,in which enhanced active constrained layer damping is carried out,and the neural network model of Genetic Algorithm-Back Propagation(GA-BP)algorithm is *** suppression and structural optimization of the SACLD manipulator model are studied based on vibration mode and damping *** modal responses of the SACLD manipulator model at rest and rotation are *** addition,the four model indices are optimized using the GA-BP neural network:axial incision size,axial incision position,circumferential incision size,and circumferential incision ***,the best model for vibration suppression is obtained.

关键词： Machine learning Vibration suppression Flexible manipulator Segmented active constrained layer damping Structural optimization

Distributed Nash Equilibrium Seeking on Compact Action Sets over Jointly Strongly Connected Switching Networks

在线全文

学校读者我要写书评

暂无评论

Journal of Systems science & Complexity 2024年第1期37卷 63-81页

作者： HE Xiongnan HUANG Jie department of Mechanical and Automation engineering The Chinese University of Hong KongHong Kong 999077China.

This paper studies the distributed Nash equilibrium seeking(DNES)problem for games whose action sets are compact and whose network graph is switching satisfying the jointly strongly connected *** keep the actions of all players in their action sets all the time,one has to resort to the projected gradient-based *** the assumption that the unique Nash equilibrium is the unique equilibrium of the pseudogradient system,an algorithm is proposed that is able to exponentially find the Nash ***,the authors also consider the distributed Nash equilibrium seeking problem for games whose actions are governed by high-order integrator dynamics and belong to some compact *** examples are used to illustrate the proposed approach.

关键词： Compact action sets jointly strongly connected switching graphs Nash equilibrium seeking projected gradient-based algorithm.

WirelessLLM:Empowering Large Language Models Towards Wireless Intelligence

在线全文

学校读者我要写书评

暂无评论

Journal of Communications and Information Networks 2024年第2期9卷 99-112页

作者： Jiawei Shao Jingwen Tong Qiong Wu Wei Guo Zijian Li Zehong Lin Jun Zhang department of Electronic and computer engineering The Hong Kong University of Science and TechnologyHong Kong 999077China

The rapid evolution of wireless technologies and the growing complexity of network infrastructures necessitate a paradigm shift in how communication networks are designed,configured,and managed. Recent advancements in large language models (LLMs) have sparked interest in their potential to revolutionize wireless communication systems. However, existing studies on LLMs for wireless systems are limited to a direct application for telecom language understanding. To empower LLMs with knowledge and expertise in the wireless domain, this paper proposes WirelessLLM, a comprehensive framework for adapting and enhancing LLMs to address the unique challenges and requirements of wireless communication networks. We first identify three foundational principles that underpin WirelessLLM:knowledge alignment, knowledge fusion, and knowledge evolution. Then,we investigate the enabling technologies to build WirelessLLM, including prompt engineering, retrieval augmented generation, tool usage, multi-modal pre-training, and domain-specific fine-tuning. Moreover, we present three case studies to demonstrate the practical applicability and benefits of WirelessLLM for solving typical problems in wireless networks. Finally, we conclude this paper by highlighting key challenges and outlining potential avenues for future research.

关键词： large language models multi-modal models wireless communications power allocation spectrum sensing protocol understanding

SWG: an architecture for sparse weight gradient computation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第2期67卷 302-321页

作者： Weiwei WU Fengbin TU Xiangyu LI Shaojun WEI Shouyi YIN School of Integrated Circuits Tsinghua University department of Electronic and computer engineering The Hong Kong University of Science and Technology

On-device training for deep neural networks(DNN) has become a trend due to various user preferences and scenarios. The DNN training process consists of three phases, feedforward(FF), backpropagation(BP), and weight gradient(WG) update. WG takes about one-third of the computation in the whole training process. Current training accelerators usually ignore the special computation property of WG and process it in a way similar to FF/BP. Besides, the extensive data sparsity existing in WG, which brings opportunities to save computation, is not well explored. Nevertheless, exploiting the optimization opportunities would meet three underutilization problems, which are caused by(1) the mismatch between WG data dimensions and hardware parallelism,(2) the full sparsity, i.e., the sparsity of feature map(Fmap),error map(Emap), and gradient, and(3) the workload imbalance resulting from irregular sparsity. In this paper, we propose a specific architecture for sparse weight gradient(SWG) computation. The architecture is designed based on hierarchical unrolling and sparsity-aware(HUSA) dataflow to exploit the optimization opportunities of the special computation property and full data sparsity. In HUSA dataflow, the data dimensions are unrolled hierarchically on the hardware architecture. A valid-data trace(VDT) mechanism is embedded in the dataflow to avoid the underutilization caused by the two-sided input sparsity. The gradient is unrolled in PE to alleviate the underutilization induced by output sparsity while maintaining the data reuse opportunities. Besides, we design an intra-and inter-column balancer(IIBLC) to dynamically tackle the workload imbalance problem resulting from the irregular sparsity. Experimental results show that with HUSA dataflow exploiting the full sparsity, SWG achieves a speedup of 12.23× over state-of-the-art gradient computation architecture, Train Ware. SWG helps to improve the energy efficiency of the state-of-the-art training accelerator LNPU from

关键词： CNN training gradient computation sparsity architecture

Hierarchical crystalline-amorphous nanocomposites with high strength and large deformability enabled by elemental diffusion

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Materials science & Technology 2024年第4期171卷 150-161页

作者： Liqiang Wang Heyi Wang Xin Zhou Huangliu Fu James Utama Surjadi Shuo Qu Xu Song Rong Fan Yang Lu department of Mechanical engineering City University of Hong KongHong KongChina Nano-Manufacturing Laboratory(NML) Shenzhen Research Institute of City University of Hong KongShenzhen 518057China Chengdu Research Institute City University of Hong KongChengdu 610200China Shenyang National Laboratory for Materials science Institute of Metal ResearchChinese Academy of SciencesShenyang 110016China department of Mechanical and Automation engineering The Chinese University of Hong KongHong KongChina department of Mechanical engineering The University of Hong KongHong KongChina

Amorphous/nanocrystalline dual-phase structures have recently emerged as an effective way for over-coming the strength-ductility trade-offand breaking the limitation of the reverse Hall-Petch ***,we proposed a new strategy to develop a hierarchical and interconnected amorphous-crystalline nanocomposite arising from the nanoscale elemental interdiffusion and oxygen adsorption behavior dur-ing thermal treatment *** nanocomposite consisted of a three-dimensional(3D)hierarchical network structure where the crystalline phase(Cr-Co-Ni-Al)was embedded into the Al-O-based amor-phous phase network with critical feature sizes encompassing three orders of magnitude(from microm-eter to nanometer scale).It can achieve ultrahigh compression yield strength of-3.6 GPa with large homogeneous deformation of over 50%*** massive interstitial atoms induced lattice distortion and hierarchical amorphous phase boundary contributed to the strength *** situ Uniaxial compression inside a transmission electron microscope(TEM)revealed that the exceptional deformability of the nanocomposites resulted from the homogenous plastic flow of nano-sized amorphous phase and the plastic co-deformation behavior restricted by the nano-architected dual-phase *** proposed dual-phase synthesis approach can outperform conventional nanolaminates design strategies in terms of the mechanical properties achievable while providing a pathway to easily tune the microstructure of these nanolaminates.

关键词： Nanolaminates Dual-phase nanocomposites In situ transmission electron microscopy Multi-component alloy

MAUN:Memory-Augmented Deep Unfolding Network for Hyperspectral Image Reconstruction

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第5期11卷 1139-1150页

作者： Qian Hu Jiayi Ma Yuan Gao Junjun Jiang Yixuan Yuan Electronic Information School Wuhan UniversityWuhan 430072China School of computer science and Technology Harbin Institute of TechnologyHarbin 150001China department of Electronic engineering Chinese University of Hong KongHong Kong 999077China

Spectral compressive imaging has emerged as a powerful technique to collect the 3D spectral information as 2D *** algorithm for restoring the original 3D hyperspectral images(HSIs)from compressive measurements is pivotal in the imaging *** approaches painstakingly designed networks to directly map compressive measurements to HSIs,resulting in the lack of interpretability without exploiting the imaging *** some recent works have introduced the deep unfolding framework for explainable reconstruction,the performance of these methods is still limited by the weak information transmission between iterative *** this paper,we propose a Memory-Augmented deep Unfolding Network,termed MAUN,for explainable and accurate HSI ***,MAUN implements a novel CNN scheme to facilitate a better extrapolation step of the fast iterative shrinkage-thresholding algorithm,introducing an extra momentum incorporation step for each iteration to alleviate the information ***,to exploit the high correlation of intermediate images from neighboring iterations,we customize a cross-stage transformer(CSFormer)as the deep denoiser to simultaneously capture self-similarity from both in-stage and cross-stage features,which is the first attempt to model the long-distance dependencies between iteration *** experiments demonstrate that the proposed MAUN is superior to other state-of-the-art methods both visually and *** code is publicly available at https://***/HuQ1an/MAUN.

关键词： Deep folding iteration