检索结果-内蒙古大学图书馆

AIF-SFDA: Autonomous Information Filter-driven Source-Free Domain Adaptation for Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Haojin Li, Heng Chen, Jianyu Zhong, Rihan Niu, Ke Fu, Huazhu Liu, Jiang Research Institute of Trustworthy Adaptive Systems Southern University of Science and Technology China Department of Computer Science and Engineering Southern University of Science and Technology China Beijing Information Science & Technology University China Institute of High Performance Computing Agency for Science Technology and Research Singapore

Decoupling domain-variant information (DVI) from domain-invariant information (DII) serves as a prominent strategy for mitigating domain shifts in the practical implementation of deep learning algorithms. However, in medical settings, concerns surrounding data collection and privacy often restrict access to both training and test data, hindering the empirical decoupling of information by existing methods. To tackle this issue, we propose an Autonomous Information Filter-driven Source-free Domain Adaptation (AIF-SFDA) algorithm, which leverages a frequency-based learnable information filter to autonomously decouple DVI and DII. Information Bottleneck (IB) and Self-supervision (SS) are incorporated to optimize the learnable frequency filter. The IB governs the information flow within the filter to diminish redundant DVI, while SS preserves DII in alignment with the specific task and image modality. Thus, the autonomous information filter can overcome domain shifts relying solely on target data. A series of experiments covering various medical image modalities and segmentation tasks were conducted to demonstrate the benefits of AIF-SFDA through comparisons with leading algorithms and ablation studies. The code is available at https://***/JingHuaMan/AIF-SFDA. Copyright © 2025, The Authors. All rights reserved.

关键词： Image segmentation

Ni1+δ metallic state and purely Ni−dx2−y2-occupied Fermi surface in SrNiO2-based superlattices at ambient pressure

学校读者我要写书评

暂无评论

Physical Review Materials 2025年第4期9卷 044802-044802页

作者： Miao Li Zhenyu Ding Yuqiang Liu Liangyu Li Gang Wu Xiaoping Yang High Magnetic Field Laboratory HFIPS Chinese Academy of Sciences Hefei 230031 China Science Island Branch of Graduate School University of Science and Technology of China Hefei 230026 China Institute of High Performance Computing (IHPC) Agency for Science Technology and Research (ASTAR) 1 Fusionopolis Way #16-16 Connexis Singapore 138632 Singapore

The rich physics of nickelates has garnered widespread attention, particularly for their superconductivity and the prominent Ni−dx2−y2-dominated Fermi surface. Motivated by the Ni1+δ superconductivity in infinite-layer NdNiO2 (Tc=15K) and high-pressure superconducting La3Ni2O7 (Tc=80K), we designed a (SrNiO2)2/(LaTiO3)1 superlattice with a two-dimensional confined bilayer NiO2. LaTiO3 acts as an electron donor, transferring a full electron to the NiO2 double layers due to the electronegativity difference between Ni(1.91) and Ti(1.54). This charge transfer transforms the Fermi surface from a multiorbital character, resembling that of NdNiO2, into one dominated purely by Ni1+δ−dx2−y2 orbitals. Factors that limited the transition temperature in NdNiO2, such as 5d−3d hybridization, self-doping between Nd and Ni atoms, and the weak p−d hybridization between Ni and O atoms, are effectively suppressed and enhanced, respectively. The nearest-neighbor eg hoppings closely resemble those in undoped infinite-layer NdNiO2, though they are smaller than those found in high-pressure La3Ni2O7. Similar to La3Ni2O7, this superlattice exhibits a low-spin magnetic state with instability stemming from nearly degenerate magnetic ground states. Additionally, F doping in NiO2 double layers can be used to fine tune the Ni valence state and the magnetic coupling, simulating the effects of oxygen pressure in the preparation. Our research offers new insights into the mechanisms driving high-Tc superconductivity in nickelates under ambient pressure, emphasizing the pivotal role of heterostructure engineering in these processes.

关键词： Fermi surface

Memory Prefetching Evaluation of Scientific Applications on A Modern HPC Arm-based Processor

学校读者我要写书评

暂无评论

IEEE Access 2025年 13卷 85898-85926页

作者： Ho, Nam Falquez, Carlos Portero, Antoni Suarez, Estela Pleiter, Dirk Forschungszentrum Jülich GmbH Jülich Supercomputing Centre Institute for Advanced Simulation Jülich Germany Computer Science Department Rhenish Friedrich Wilhelm University of Bonn Bonn Germany SiPEARL Maisons-Laffitte France KTH Royal Institute of Technology PDC Center for High Performance Computing Stockholm Sweden University of Groningen Bernoulli Institute FSE Department of Computer Science Groningen Netherlands

Memory prefetching is a well-known technique for mitigating the negative impact of memory access latencies on memory bandwidth. This problem has become more pressing as improvements in memory bandwidth have not kept pace with increases in computational power. While much existing work has been devoted to finding appropriate prefetching techniques for specific workloads, few provide insight into the behavior of scientific applications to better understand the impact of prefetchers. This paper investigates the impact of hardware prefetchers on the latest Arm-based high-end processor architectures. In this work, we investigate memory access patterns by analyzing locality properties and visualizing delta and repetitive address patterns. A deeper understanding of memory access patterns allows the use of the appropriate prefetcher and reaching a better correlation between access pattern properties and prefetcher performance. This can guide future co-design efforts. We evaluated traditional and innovative prefetchers using a gem5-based model of Arm Neoverse V1 cores. The model features a 16-core architecture, using Amazon's Graviton 3 processor as a hardware reference, but substituting DDR5 by high bandwidth memory (HBM2). We performed a detailed prefetching evaluation focusing on stencil, sparse matrix-vector multiplication, and Breadth-First Search kernels. These kernels represent a broad range of the applications running on today's high-performance computing (HPC) systems, which are sensitive to memory performance. © 2013 IEEE.

关键词： computer simulation high performance computing Memory prefetcher

Risk-mediated dynamic regulation of effective contacts de-synchronizes outbreaks in metapopulation epidemic models

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zunker, Henrik Dönges, Philipp Lenz, Patrick Contreras, Seba Kühn, Martin J. Institute of Software Technology Department of High-Performance Computing German Aerospace Center Cologne Germany Max Planck Institute for Dynamics and Self-Organization Göttingen Germany Institute for the Dynamics of Complex Systems University of Göttingen Göttingen Germany Life and Medical Sciences Institute Bonn Center for Mathematical Life Sciences University of Bonn Bonn Germany

Metapopulation epidemic models help capture the spatial dimension of infectious disease spread by dividing heterogeneous populations into separate but interconnected communities, represented by nodes in a network. In the event of an epidemic, an important research question is, to what degree spatial information (i.e., regional or national) is relevant for mitigation and (local) policymakers. This study investigates the impact of different levels of information on nationwide epidemic outcomes, modeling the reaction to the measured hazard as a feedback loop reducing contact rates in a metapopulation model based on ordinary differential equations (ODEs). Using COVID-19 and high-resolution mobility data for Germany of 2020 as a case study, we found two markedly different regimes depending on the maximum contact reduction ψmax: mitigation and suppression. In the regime of (modest) mitigation, gradually increasing ψmax from zero to moderate levels delayed and spread out the onset of infection waves while gradually reducing the peak values. This effect was more pronounced when the contribution of regional information was low relative to national data. In the suppression regime, the feedback-induced contact reduction is strong enough to extinguish local outbreaks and decrease the mean and variance of the peak day distribution, thus regional information was more important. When suppression or elimination is impossible, ensuring local epidemics are desynchronized helps to avoid hospitalization or intensive care bottlenecks by reallocating resources from less-affected *** Codes 3404, 9210, 92B05 Copyright © 2025, The Authors. All rights reserved.

关键词： COVID-19

Vision-Language Model IP Protection via Prompt-based Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Lianyu Wang, Meng Fu, Huazhu Zhang, Daoqiang The Key Laboratory of Brain-Machine Intelligence Technology Ministry of Education China Centre for Innovation and Precision Eye Health Yong Loo Lin School of Medicine National University of Singapore Singapore Institute of High Performance Computing Agency for Science Technology and Research Singapore

Vision-language models (VLMs) like CLIP (Contrastive Language-Image Pre-Training) have seen remarkable success in visual recognition, highlighting the increasing need to safeguard the intellectual property (IP) of well-trained models. Effective IP protection extends beyond ensuring authorized usage;it also necessitates restricting model deployment to authorized data domains, particularly when the model is fine-tuned for specific target domains. However, current IP protection methods often rely solely on the visual backbone, which may lack sufficient semantic richness. To bridge this gap, we introduce IP-CLIP, a lightweight IP protection strategy tailored to CLIP, employing a prompt-based learning approach. By leveraging the frozen visual backbone of CLIP, we extract both image style and content information, incorporating them into the learning of IP prompt. This strategy acts as a robust barrier, effectively preventing the unauthorized transfer of features from authorized domains to unauthorized ones. Additionally, we propose a style-enhancement branch that constructs feature banks for both authorized and unauthorized domains. This branch integrates self-enhanced and cross-domain features, further strengthening IP-CLIP’s capability to block features from unauthorized domains. Finally, we present new three metrics designed to better balance the performance degradation of authorized and unauthorized domains. Comprehensive experiments in various scenarios demonstrate its promising potential for application in IP protection tasks for VLMs. © 2025, CC BY-NC-ND.

关键词： Semantics

Optimized circuits for windowed modular arithmetic with applications to quantum attacks against RSA

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Luongo, Alessandro Narasimhachar, Varun Sireesh, Adithya Centre for Quantum Technologies National University of Singapore Singapore Inveriant Pte. Ltd. Singapore Institute of High Performance Computing Agency for Science Technology and Research Singapore School of Informatics University of Edinburgh Scotland United Kingdom

Windowed arithmetic [Gidney, 2019] is a technique for reducing the cost of quantum arithmetic circuits with space–time tradeoffs using memory queries to precomputed tables. It can reduce the asymptotic cost of modular exponentiation from O(n3) to O(n3/log2 n) operations, resulting in the current state-of-the-art compilations of quantum attacks against modern cryptography. In this work we introduce four optimizations to windowed modular exponentiation. We (1) show how the cost of unlookups can be reduced by 66% asymptotically in the number of bits, (2) illustrate how certain addresses can be bypassed, reducing both circuit depth and the overall lookup cost, (3) demonstrate that multiple lookup–addition operations can be merged into a single, larger lookup at the start of the modular exponentiation circuit, and (4) reduce the depth of the unary conversion for unlookups. On a logical level, this leads to a 3% improvement in Toffoli count and Toffoli depth for modular exponentiation circuits relevant to cryptographic applications. This translates to some improvements on [Gidney and Ekerå, 2021]’s factoring algorithm: for a given number of physical qubits, our improvements show a reduction in the expected runtime from 2% to 6% for factoring RSA-2048 *** Codes 81P68, 68Q09, 68Q12 © 2025, CC0.

关键词： Table lookup

Ad2Mix: Adversarial and Adaptive Mixup for Unsupervised Domain Adaptation

学校读者我要写书评

暂无评论

Ad2Mix: Adversarial and Adaptive Mixup for Unsupervised Doma...

IEEE Workshop on Applications of Computer Vision (WACV)

作者： Lei Zhu Yanyu Xu Yong Liu Rick Siow Mong Goh Xinxing Xu Technology and Research (A*STAR) Institute of High Performance Computing (IHPC) Agency for Science Singapore Republic of Singapore The Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR) Shandong University Shandong China Microsoft Research Asia Singapore

ISBN: (数字)9798331510831

ISBN: (纸本)9798331510848

Transformer has recently gained tremendous popularity in unsupervised domain adaptation tasks due to its superior generalization ability. State-of-the-art methods leverage mixup to build an intermediate domain to reduce domain gap. However, such strategy becomes less effective when the domain gap becomes large, as the domain gap between intermediate domain and source domain is not minimized and the constructed intermediate domain is non informative. How to address the adaptation problem when domain gap becomes large is an important research problem in domain adaptation. In this paper, we propose an adversarial and adaptive mixup (Ad 2 mix) framework which gradually aligns the intermediate domain towards source domain to fully unleash the potential of both the transformer architecture and mixup to address the large domain gap problem. Specifically, we formulate a general framework for intermediate domain learning with mixup. We propose adversarial mixup with a specially designed mixup alike adversarial adaptation operation to reduce the domain gap between the intermediate domain and source domain. To construct an informative intermediate domain, unlike existing methods which utilize a Beta distribution to generate mixup coefficients to interpolate source and target data, we adaptively assign mixup coefficient for each target data instance based on their transferability and discriminativity information. Our framework creates a natural curriculum of intermediate domains from near source domain to near target domain for gradual adaptation. Extensive experimental studies and evaluations on three public domain adaptation benchmark datasets and one medical domain adaptation task demonstrate the superiority of our framework.

关键词： Computer vision Knowledge based systems Machine learning Benchmark testing Transformers

LLM Knows Geometry Better than Algebra: Numerical Understanding of LLM-Based Agents in A Trading Arena

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ma, Tianmi Du, Jiawei Huang, Wenxin Wang, Wenjie Xie, Liang Zhong, Xian Zhou, Joey Tianyi Hubei Key Laboratory of Transportation Internet of Things Wuhan University of Technology China Hubei Key Laboratory of Big Data Intelligent Analysis and Application Hubei University China Centre for Frontier AI Research Agency for Science Technology and Research Singapore Institute of High Performance Computing Agency for Science Technology and Research Singapore School of Computing National University of Singapore Singapore School of Science Wuhan University of Technology China

Recent advancements in large language models (LLMs) have significantly improved performance in natural language processing tasks. However, their ability to generalize to dynamic, unseen tasks, particularly in numerical reasoning, remains a challenge. Existing benchmarks mainly evaluate LLMs on problems with predefined optimal solutions, which may not align with real-world scenarios where clear answers are absent. To bridge this gap, we design the Agent Trading Arena, a virtual numerical game simulating complex economic systems through zero-sum games, where agents invest in stock portfolios. Our experiments reveal that LLMs, including GPT-4o, struggle with algebraic reasoning when dealing with plain-text stock data, often focusing on local details rather than global trends. In contrast, LLMs perform significantly better with geometric reasoning when presented with visual data, such as scatter plots or K-line charts, suggesting that visual representations enhance numerical reasoning. This capability is further improved by incorporating the reflection module, which aids in the analysis and interpretation of complex data. We validate our findings on NASDAQ STOCK dataset, where LLMs demonstrate stronger reasoning with visual data compared to text. Our code and data are publicly available at https://***/wekjsdvnm/***. © 2025, CC BY.

关键词： Natural language processing systems

Fundus Image Quality Assessment and Enhancement: a Systematic Review

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Heng Li, Haojin Ou, Mingyang Yu, Xiangyang Zhang, Xiaoqing Niu, Ke Fu, Huazhu Liu, Jiang Research Institute of Trustworthy Autonomous Systems SUSTech Shenzhen China Department of Computer Science and Engineering SUSTech Shenzhen China Center for High Performance Computing and Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Computer School Beijing Information Science and Technology University Beijing China Singapore

As an affordable and convenient eye scan, fundus photography holds the potential for preventing vision impairment, especially in resource-limited regions. However, fundus image degradation is common under intricate imaging environments, impacting following diagnosis and treatment. Consequently, image quality assessment (IQA) and enhancement (IQE) are essential for ensuring the clinical value and reliability of fundus images. While existing reviews offer some overview of this field, a comprehensive analysis of the interplay between IQA and IQE, along with their clinical deployment challenges, is lacking. This paper addresses this gap by providing a thorough review of fundus IQA and IQE algorithms, research advancements, and practical applications. We outline the fundamentals of the fundus photography imaging system and the associated interferences, and then systematically summarize the paradigms in fundus IQA and IQE. Furthermore, we discuss the practical challenges and solutions in deploying IQA and IQE, as well as offer insights into potential future research directions. Copyright © 2025, The Authors. All rights reserved.

关键词： Photography