检索结果-内蒙古大学图书馆

Mamba-Sea: A Mamba-based Framework with Global-to-Local Sequence Augmentation for Generalizable Medical Image Segmentation

引用

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Cheng, Zihan Guo, Jintao Zhang, Jian Qi, Lei Zhou, Luping Shi, Yinghuan Gao, Yang Shanghai Jiao Tong University School of Medicine Shanghai200025 China Nanjing University National Institute of Healthcare Data Science Jiangsu Nanjing210093 China Nanjing University State Key Laboratory for Novel Software Technology National Institute of Healthcare Data Science Jiangsu Nanjing210093 China Southeast University School of Computer Science and Engineering Key Lab of Computer Network and Information Integration Jiangsu Nanjing211189 China The University of Sydney School of Electrical and Information Engineering SydneyNSW2006 Australia

To segment medical images with distribution shifts, domain generalization (DG) has emerged as a promising setting to train models on source domains that can generalize to unseen target domains. Existing DG methods are mainly based on CNN or ViT architectures. Recently, advanced state space models, represented by Mamba, have shown promising results in various supervised medical image segmentation. The success of Mamba is primarily owing to its ability to capture long-range dependencies while keeping linear complexity with input sequence length, making it a promising alternative to CNNs and ViTs. Inspired by the success, in the paper, we explore the potential of the Mamba architecture to address distribution shifts in DG for medical image segmentation. Specifically, we propose a novel Mamba-based framework, Mamba-Sea, incorporating global-to-local sequence augmentation to improve the model’s generalizability under domain shift issues. Our Mamba-Sea introduces a global augmentation mechanism designed to simulate potential variations in appearance across different sites, aiming to suppress the model’s learning of domain-specific information. At the local level, we propose a sequence-wise augmentation along input sequences, which perturbs the style of tokens within random continuous sub-sequences by modeling and resampling style statistics associated with domain shifts. To our best knowledge, Mamba-Sea is the first work to explore the generalization of Mamba for medical image segmentation, providing an advanced and promising Mamba-based architecture with strong robustness to domain shifts. Remarkably, our proposed method is the first to surpass a Dice coefficient of 90% on the Prostate dataset, which exceeds previous SOTA of 88.61%. © 1982-2012 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Ultra-Long 1D Double Perovskite CsAg0.4Cu1.6I3 Single Crystal for UV Imaging Memory

引用

Advanced Functional Materials 2025年

作者： Cao, Fa Liu, Ying Hong, Enliu Xing, Ruofei Han, Wushuang Han, Sancan Sun, Bin Nanjing210023 China Department of Materials Science State Key Laboratory of Molecular Engineering of Polymers Fudan University Shanghai200433 China School of Opto-electronic and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China

Emerging copper-based lead-free perovskites have garnered significant attention for ultraviolet (UV) photodetection due to their atmospheric stability and optoelectronic properties. However, their practical applications are hindered by low carrier density and limited device performance. In this work, a breakthrough is reported in synthesizing ultra-long 1D double perovskite CsAgxCu2-xI3 (0≤x≤2) single crystals (SCs) via an anti-solvent recrystallization method. The SCs achieve unprecedented lengths of 2.2 cm for CsAg0.4Cu1.6I3 and 3 cm for CsAg2I3 (CAI123), setting new benchmarks for Cu/Ag-based perovskites. By tuning the Ag-doping ratio, the bandgap is continuously modulated from 3.60 to 3.24 eV, enabling tailored optoelectronic responses. The optimized CsAg0.4Cu1.6I3-based photodetector exhibits exceptional performance at 360 nm, with a peak responsivity of 82.5 mA W−1 and a high detectivity of 1.3 × 1012 Jones (@ 3V;light intensity: 2.8 mW cm−2), superior than CsCu2I3 (20.7 mA W−1) and CAI123 (15.8 mA W−1). Notably, the detector array demonstrates pioneering UV imaging memory capabilities, retaining patterns for over 10 s via slow carrier release from self-trapped and defect-related states. This work not only advances the synthesis of eco-friendly perovskites but also unlocks novel functionalities for next-generation optoelectronics in UV sensing, imaging, and memory devices. © 2025 Wiley-VCH GmbH.

关键词： Photodetectors

来源：评论

学校读者我要写书评

暂无评论

Machine learning automation

引用

National science Review 2024年第8期11卷 7-8页

作者： Zongben Xu Zhi-Hua Zhou Wenwu Zhu School of Mathematics and Statistics Xi'an Jiaotong University National Key Laboratory for Novel Software Technology Nanjing University Department of Computer Science and Technology Tsinghua University

With the exponential growth of big data and advancements in large-scale foundation model techniques, the field of machine learning has embarked on an unprecedented golden era. This period is characterized by significant innovations across various aspects of machine learning, including data exploitation, network architecture development, loss function settings and algorithmic innovation.

关键词： automation learning machine

来源：评论

学校读者我要写书评

暂无评论

Cooperative Co-evolutionary Artificial Bee Colony Algorithm Based on Hierarchical Communication Model

引用

Chinese Journal of Electronics 2016年第3期25卷 570-576页

作者： HU Chunyu LIU Hong ZHANG Peng School of Information Science and Engineering Shandong Normal University Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technology

Canonical Artificial bee colony(ABC) algorithm with a single species is insufficient to extend the diversity of solutions and may be trapped into the local optimal solution. This paper proposes a new co-evolutionary ABC algorithm(HABC) based on Hierarchical communication model(HCM). HCM combines advantages of global and local communication pattern. With adjustment strategies on species and groups, HCM can reduce the computational complexity dynamically. Performance tests show that the HABC algorithm exhibit good performance on accuracy, robustness and convergence speed. Compared with ABC and Integrated co-evolution algorithm(IABC),HABC performs better in solving complex multimodal functions.

关键词： Artificial bee colony(ABC) algorithm Multi-species co-evolution Hierarchical communication model(HCM) Dynamical adjustment

来源：评论

学校读者我要写书评

暂无评论

Model based verification of dynamically evolvable service oriented systems

引用

science China(Information sciences) 2016年第3期59卷 5-21页

作者： Yu ZHOU Jidong GE Pengcheng ZHANG Weigang WU College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics State Key Laboratory for Novel Software Technology Nanjing University College of Computer and Information Hohai University College of Computer Science Sun Yat-Sen University

Dynamic evolution is highly desirable for service oriented systems in open environments. For the evolution to be trusted, it is crucial to keep the process consistent with the specification. In this paper, we study two kinds of evolution scenarios and propose a novel verification approach based on hierarchical timed automata to model check the underlying consistency with the specification. It examines the procedures before, during and after the evolution process, respectively and can support the direct modeling of temporal aspects, as well as the hierarchical decomposition of software structures. Probabilities are introduced to model the uncertainty characterized in open environments and thus can support the verification of parameter-level evolution. We present a flattening algorithm to facilitate automated verification using the mainstream timed automata based model checker –UPPAAL(integrated with UPPAAL-SMC). We also provide a motivating example with performance evaluation that complements the discussion and demonstrates the feasibility of our approach.

关键词： dynamic evolution verification model checking service oriented systems timed automata

来源：评论

学校读者我要写书评

暂无评论

IO dependent SSD cache allocation for elastic Hadoop applications

引用

science China(Information sciences) 2018年第5期61卷 55-71页

作者： Zhen TANG Wei WANG Lei SUN Yu HUANG Heng WU Jun WEI Tao HUANG State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences Tianjin Massive Data Processing Technology Laboratory Tianjin Shenzhou General Data Technology Co. Ltd. State Key Laboratory for Novel Software Technology Nanjing University

Elastic Hadoop applications consisting of multiple virtual machines(VMs) are widely used to support big data analysis and processing. In this scenario, flash-based solid state drive(SSD) is usually deployed on hypervisors and used as the cache to improve the IO performance. However, existing SSD caching schemes are mostly VM-centric, which focus on the low-level IO performance metrics of individual VMs. They may not lead to the optimized performance of elastic Hadoop applications, i.e., the job completion time(JCT), as the importance of VMs inside the application are different even though they have the similar low-level IO patterns. Considering the IO dependency among VMs and figuring out the importance, which we regard as the application-centric metrics, may potentially better improve the performance. We present IO dependency based requirement model, to characterize the requirement of SSD cache for each VM inside the elastic Hadoop application, and then use it in a genetic algorithm(GA) based approach to calculate the nearly optimal weights of VMs for allocating the per-VM SSD cache space and the capacity of the I/O operations per second(IOPS). Furthermore, we present a tool AC-SSD based on the approach and introduce the closed-loop adaptation to react to continuously changing workloads. The evaluation shows that by using AC-SSD, the JCT is reduced by up to 39% for IO sensitive workloads, up to 29% for continuously changing workloads, and over 12.5% for different scale of data comparing to the shared cache.

关键词： Hadoop SSD cache resource management virtualization

来源：评论

学校读者我要写书评

暂无评论

Rendering discrete participating media using geometrical optics approximation

引用

Computational Visual Media 2022年第3期8卷 425-444页

作者： Jie Guo Bingyang Hu Yanjun Chen Yuanqi Li Yanwen Guo Ling-Qi Yan State Key Lab for Novel Software Technology Nanjing UniversityNanjing 210023China Department of Computer Science UC Santa BarbaraUSA

We consider the scattering of light in participating media composed of sparsely and randomly distributed discrete *** particle size is expected to range from the scale of the wavelength to several orders of magnitude greater,resulting in an appearance with distinct graininess as opposed to the smooth appearance of continuous *** fundamental issue in the physically-based synthesis of such appearance is to determine the necessary optical properties in every local *** these properties vary spatially,we resort to geometrical optics approximation(GOA),a highly efficient alternative to rigorous Lorenz–Mie theory,to quantitatively represent the scattering of a single *** enables us to quickly compute bulk optical properties for any particle size *** then use a practical Monte Carlo rendering solution to solve energy transfer in the discrete participating *** proposed framework is the first to simulate a wide range of discrete participating media with different levels of graininess,converging to the continuous media case as the particle concentration increases.

关键词： light scattering geometrical optics approximation(GOA) discrete participating media volume rendering

来源：评论

学校读者我要写书评

暂无评论

Modeling Topic-Based Human Expertise for Crowd Entity Resolution

引用

Journal of computer science & technology 2018年第6期33卷 1204-1218页

作者： Sai-Sai Gong Wei Hu Wei-Yi Ge Yu-Zhong Qu State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210023 China Science and Technology on Information Systems Engineering Laboratory Nanjing 210007 China

Entity resolution （ER） aims to identify whether two entities in an ER task refer to the same real-world *** ER uses humans, in addition to machine algorithms, to obtain the truths of ER tasks. However, inaccurate orerroneous results are likely to be generated when humans give unreliable judgments. Previous studies have found thatcorrectly estimating human accuracy or expertise in crowd ER is crucial to truth inference. However, a large number ofthem assume that humans have consistent expertise over all the tasks, and ignore the fact that humans may have variedexpertise on different topics （e.g., music versus sport）. In this paper, we deal with crowd ER in the Semantic Web *** identify multiple topics of ER tasks and model human expertise on different topics. Furthermore, we leverage similartask clustering to enhance the topic modeling and expertise estimation. We propose a probabilistic graphical model thatcomputes ER task similarity, estimates human expertise, and infers the task truths in a unified framework. Our evaluationresults on real-world and synthetic datasets show that, compared with several state-of-the-art approaches, our proposedmodel achieves higher accuracy on the task truth inference and is more consistent with the human real expertise.

关键词： entity resolution crowdsourcing human expertise topic modeling task similarity

来源：评论

学校读者我要写书评

暂无评论

LimTDD: A Compact Decision Diagram Integrating Tensor and Local Invertible Map Representations

arXiv

引用

arXiv 2025年

作者： Hong, Xin Dai, Aochu Gao, Dingchao Li, Sanjiang Ji, Zhengfeng Ying, Mingsheng and State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences China Department of Computer Science and Technology Tsinghua University China Centre for Quantum Software and Information University of Technology Sydney Australia

Tensor Decision Diagrams (TDDs) provide an efficient structure for representing tensors by combining techniques from both tensor networks and decision diagrams, demonstrating competitive performance in quantum circuit simulation and verification. However, existing decision diagrams, including TDDs, fail to exploit isomorphisms within tensors, limiting their compression efficiency. This paper introduces Local Invertible Map Tensor Decision Diagrams (LimTDDs), an extension of TDD that integrates local invertible maps (LIMs) to achieve more compact representations. Unlike LIMDD, which applies Pauli operators to quantum states, LimTDD generalizes this approach using the XP-stabilizer group, enabling broader applicability. We develop efficient algorithms for normalization and key tensor operations, including slicing, addition, and contraction, essential for quantum circuit simulation and verification. Theoretical analysis shows that LimTDD surpasses TDD in compactness while maintaining its generality and offers exponential advantages over both TDD and LIMDD in the best-case scenarios. Experimental results validate these improvements, demonstrating LimTDD’s superior efficiency in quantum circuit simulation and functionality computation. Copyright © 2025, The Authors. All rights reserved.

关键词： Quantum electronics

来源：评论

学校读者我要写书评

暂无评论

Co-metric： a metric learning algorithm for data with multiple views

引用

Frontiers of computer science 2013年第3期7卷 359-369页

作者： Qiang QIAN Songcan CHEN Department of Computer Science and Engineering Nanjing University of Aeronautics and Astronautics Nanjing 210016 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093 China

We address the problem of metric learning for multi-view data. Many metric learning algorithms have been proposed, most of them focus just on single view circumstances, and only a few deal with multi-view data. In this paper, motivated by the co-training framework, we propose an algorithm-independent framework, named co-metric, to learn Mahalanobis metrics in multi-view settings. In its implementation, an off-the-shelf single-view metric learning algorithm is used to learn metrics in individual views of a few labeled examples. Then the most confidently-labeled examples chosen from the unlabeled set are used to guide the metric learning in the next loop. This procedure is repeated until some stop criteria are met. The framework can accommodate most existing metric learning algorithms whether types-of- side-information or example-labels are used. In addition it can naturally deal with semi-supervised circumstances under more than two views. Our comparative experiments demon- strate its competiveness and effectiveness.

关键词： multi-view learning metric learning algorithm- independent framework

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：