检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Yeo, Joel Loh, N. Duane Paniagua-Dominguez, Ramon Kuznetsov, Arseniy I. NUS Graduate School for Integrative Sciences and Engineering Programme National University of Singapore Singapore119077 Singapore Department of Physics National University of Singapore Singapore117551 Singapore Department of Biological Sciences National University of Singapore Singapore117558 Singapore 2 Fusionopolis Way Innovis #08-03 Singapore138634 Singapore

The miniaturization of optics through the use of two-dimensional metalenses has enabled novel applications in imaging. To date, single-lens imaging remains the most common configuration, in part due to the limited focusing efficiency of metalenses. This results in limitations when it comes to wavefront manipulation and, thus, unavoidable aberrations in the formed image that require computational deconvolution to deblur the image. For certain lens profiles, such as the most common hyperbolic one that results in the highest efficiencies, at large fields of view, spatially-varying aberrations such as coma or astigmatism are prominent. These aberrations cannot be corrected for by traditional deconvolution methods, such as Wiener filtering. Here, we develop a spatially-varying deconvolution algorithm based on eigenvalue column-wise decomposition (eigenCWD). EigenCWD solves a minimization problem of the error between the measured image and the estimated image of the object to be reconstructed through an approximate forward blurring model. This approximate forward model uses an eigendecomposition of the spatially-varying point spread functions for fast computation, allowing for efficient scaling to larger image sizes and blurring kernels common in metalens imaging. We demonstrate eigenCWD’s ability to correct spatially-varying blur and distortions for various lens profiles, surpassing that of the Wiener filter. © 2025, CC BY-NC-SA.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Near-Sensor Edge Computing System Enabled by a CMOS Compatible Photonic Integrated Circuit Platform Using Bilayer AlN/Si Waveguides

引用

Nano-Micro Letters 2025年第1期17卷 1-20页

作者： Ren, Zhihao Zhang, Zixuan Zhuge, Yangyang Xiao, Zian Xu, Siyu Zhou, Jingkai Lee, Chengkuo Department of Electrical and Computer Engineering National University of Singapore Singapore 117583 Singapore Center for Intelligent Sensors and MEMS National University of Singapore Singapore 117608 Singapore National Centre for Advanced Integrated Photonics (NCAIP) Singapore 639798 Singapore NUS Graduate School - Integrative Sciences and Engineering Programme (ISEP) National University of Singapore 21 Lower Kent Ridge Road Singapore 119077 Singapore

A novel near-sensor edge computing system integrates aluminum nitride (AlN) microrings for photonic feature extraction and Si Mach–Zehnder interferometers for photonic neural network operations, achieving real-time artificial intelligence (AI) processing. Demonstrates high classification accuracy (96.77% for gestures, 98.31% for gaits) with low latency (< 10 ns) and minimal energy consumption (< 0.34 pJ). Enables low-power, high-speed AI applications with seamless hybrid photonic-electronic integration on a bilayer AlN/Si waveguide platform. © The Author(s) 2025.

关键词： Aluminum nitride Edge computing Neural networks Photonic integrated circuits Wearable sensors

来源：评论

学校读者我要写书评

暂无评论

Reconfigurable nonlinear photonic activation function for photonic neural network based on non-volatile opto-resistive RAM switch

引用

Light(Science & Applications) 2022年第10期11卷 2512-2522页

作者： Zefeng Xu Baoshan Tang Xiangyu Zhang Jin Feng Leong Jieming Pan Sonu Hooda Evgeny Zamburg Aaron Voon-Yew Thean Integrative Sciences and Engineering Programme NUS Graduate SchoolNational University of SingaporeSingaporeSingapore Department of Electrical and Computer Engineering National University of Singapore4 Engineering Drive 3Singapore 117583Singapore

Photonic neural network has been sought as an alternative solution to surpass the efficiency and speed bottlenecks of electronic neural *** that the integrated Mach-Zehnder Interferometer(MZI)mesh can perform vector-matrix multiplication in photonic neural network,a programmable in-situ nonlinear activation function has not been proposed to date,suppressing further advancement of photonic neural ***,we demonstrate an efficient in-situ nonlinear accelerator comprising a unique solution-processed two-dimensional(2D)MoS2 Opto-Resistive RAM Switch(ORS),which exhibits tunable nonlinear resistance switching that allow us to introduce nonlinearity to the photonic neuron which overcomes the linear voltage-power relationship of typical photonic *** reconfigurable scheme enables implementation of a wide variety of nonlinear ***,we confirm its feasibility and capability for MNIST handwritten digit recognition,achieving a high accuracy of 91.6%.Our accelerator constitutes a major step towards the realization of in-situ photonic neural network and pave the way for the integration of photonic integrated circuits(PIC).

关键词： network neural nonlinear

来源：评论

学校读者我要写书评

暂无评论

GENESHIFT: IMPACT OF DIFFERENT SCENARIO SHIFT ON JAILBREAKING LLM

arXiv

引用

arXiv 2025年

作者： Wu, Tianyi Xue, Zhiwei Liu, Yue Zhang, Jiaheng Hooi, Bryan Ng, See-Kiong Integrative Sciences and Engineering Programme NUS Graduate School National University of Singapore Singapore National University of Singapore Singapore Department of Computer Science School of Computing National University of Singapore Singapore

Jailbreak attacks, which aim to cause LLMs to perform unrestricted behaviors, have become a critical and challenging direction in AI safety. Despite achieving the promising attack success rate using dictionary-based evaluation, existing jailbreak attack methods fail to output detailed contents to satisfy the harmful request, leading to poor performance on GPT-based evaluation. To this end, we propose a black-box jailbreak attack termed GeneShift, by using a genetic algorithm to optimize the scenario shifts. Firstly, we observe that the malicious queries perform optimally under different scenario shifts. Based on it, we develop a genetic algorithm to evolve and select the hybrid of scenario shifts. It guides our method to elicit detailed and actionable harmful responses while keeping the seemingly benign facade, improving stealthiness. Extensive experiments demonstrate the superiority of GeneShift. Notably, GeneShift increases the jailbreak success rate from 0% to 60% when direct prompting alone would fail. We will open-source our code. © 2025, CC BY.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Factorial design analytics on effects of material parameter uncertainties in multiphysics modeling of additive manufacturing

引用

npj Computational Materials 2023年第1期9卷 1848-1865页

作者： Amanda Giam Fan Chen Jiaxiang Cai Wentao Yan Integrative Sciences and Engineering Programme NUS Graduate SchoolNational University of SingaporeSingapore 119077Singapore Department of Mechanical Engineering National University of SingaporeSingapore 117575Singapore Department of Industrial Systems Engineering and Management National University of SingaporeSingapore 117576Singapore

A bottleneck in Laser Powder Bed Fusion(L-PBF)metal additive manufacturing(AM)is the quality inconsistency of its *** address this issue without costly experimentation,computational multi-physics modeling has been used,but the effectiveness is limited by parameter uncertainties and their *** propose a full factorial design and variable selection approach for the analytics of main and interaction effects arising from material parameter uncertainties in multi-physics *** is collected from high-fidelity thermal-fluid simulations based on a 2-level full factorial design for 5 selected material *** physical phenomena of the L-PBF process are analyzed to extract physics-based domain knowledge,which are used to establish a validation checkpoint for our *** data visualization with half-normal probability plots,interaction plots and standard deviation plots,is used to assess if the checkpoint is being *** then apply the combination of best subset selection and the LASSO method on multiple linear regression models for comprehensive variable *** yield statistically and phyiscally validated findings with practical implications,emphasizing the importance of parameter interactions under uncertainty,and their relation to the underlying physics of L-PBF.

关键词： parameter additive analytic

来源：评论

学校读者我要写书评

暂无评论

NUS-IDS@eRisk2024: Ranking Sentences for Depression Symptoms using Early Maladaptive Schemas and Ensembles 25

NUS-IDS@eRisk2024: Ranking Sentences for Depression Symptoms...

引用

25th Working Notes of the Conference and Labs of the Evaluation Forum, CLEF 2024

作者： Ang, Beng Heng Gollapalli, Sujatha Das Ng, See-Kiong Integrative Sciences and Engineering Programme National University of Singapore Singapore Institute of Data Science National University of Singapore Singapore

In this paper, we describe the techniques developed by our team, NUS-IDS, for eRisk2024 Task 1. This task focuses on ranking and identifying sentences from social media that express symptoms of depression as outlined in the Beck's Depression Inventory-II (BDI) questionnaire. We developed five different configurations for this task using ensembles of unique sets of base predictors. The base predictors use sentence-transformer models fine-tuned with Contrastive Learning using the task's training data and expressive exemplars relevant to both the BDI symptoms and the features from the well-established taxonomy of Early Maladaptive Schemas (EMS) obtained by prompting GPT-4. Among our five submitted configurations, three of them outperformed competing runs in at least two metrics. Overall, our Configuration 5 is the best performing among all runs in three out of four competition metrics for both the Majority voting and Unanimity evaluations. We further analyzed our configurations and derive insights on the complexities of Ensemble Learning as well as the use of expressive exemplars for this task. © 2024 Copyright for this paper by its authors.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Identifying Early Maladaptive Schemas from Mental Health Question Texts

Identifying Early Maladaptive Schemas from Mental Health Que...

引用

2023 Findings of the Association for Computational Linguistics: EMNLP 2023

作者： Gollapalli, Sujatha Das Ang, Beng Heng Ng, See-Kiong Institute of Data Science National University of Singapore Singapore Integrative Sciences and Engineering Programme National University of Singapore Singapore

ISBN: (纸本)9798891760615

In Psychotherapy, maladaptive schemas- negative perceptions that an individual has of the self, others, or the world that endure despite objective reality- often lead to resistance to treatments and relapse of mental health issues such as depression, anxiety, panic attacks etc. Identification of early maladaptive schemas (EMS) is thus a crucial step during Schema Therapy-based counseling sessions, where patients go through a detailed and lengthy EMS questionnaire. However, such an approach is not practical in 'offline' counseling scenarios, such as community QA forums which are gaining popularity for people seeking mental health support. In this paper, we investigate both LLM (Large Language Models) and non-LLM approaches for identifying EMS labels using resources from Schema Therapy. Our evaluation indicates that recent LLMs can be effective for identifying EMS but their predictions lack explainability and are too sensitive to precise 'prompts'. Both LLM and non-LLM methods are unable to reliably address the null cases, i.e. cases with no EMS labels. However, we posit that the two approaches show complementary properties and together, they can be used to further devise techniques for EMS identification. © 2023 Association for Computational Linguistics.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

Socratic Question Generation: A Novel Dataset, Models, and Evaluation 17

Socratic Question Generation: A Novel Dataset, Models, and E...

引用

17th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2023

ISBN: (纸本)9781959429449

Socratic questioning is a form of reflective inquiry often employed in education to encourage critical thinking in students, or to elicit awareness of beliefs and perspectives in a subject during therapeutic counseling. Specific types of Socratic questions are employed for enabling reasoning and alternate views against the context of individual personal opinions on a topic. Socratic contexts are different from traditional question generation contexts where "answer-seeking" questions are generated against a given formal passage on a topic, narrative stories or conversations. We present SocratiQ, the first large dataset of 110K (question, context) pairs for enabling studies on Socratic Question Generation (SoQG). We provide an in-depth study on the various types of Socratic questions and present models for generating Socratic questions against a given context through prompt tuning. Our automated and human evaluation results demonstrate that our SoQG models can produce realistic, type-sensitive, human-like Socratic questions enabling potential applications in counseling and coaching. © 2023 Association for Computational Linguistics.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Grasping by Spiraling: Reproducing Elephant Movements with Rigid-Soft Robot Synergy

arXiv

引用

arXiv 2025年

作者： Huang, Huishi Wang, Haozhe Fang, Chongyu Yan, Mingge Xu, Ruochen Zhang, Yiyuan Wang, Zhanchi Ying, Fengkang Liu, Jun Laschi, Cecilia Ang, Marcelo H. Advanced Robotics Centre National University of Singapore 117608 Singapore Integrative Sciences and Engineering Programme National University of Singapore Graduate School 119077 Singapore School of Computer Science and Technology University of Science and Technology of China Anhui Hefei230027 China 138632 Singapore

—The logarithmic spiral is observed as a common pattern in several living beings across kingdoms and species. Some examples include fern shoots, prehensile tails, and soft limbs like octopus arms and elephant trunks. In the latter cases, spiraling is also used for grasping. Motivated by how this strategy simplifies behavior into kinematic primitives and combines them to develop smart grasping movements, this work focuses on the elephant trunk, which is more deeply investigated in the literature. We present a soft arm combined with a rigid robotic system to replicate elephant grasping capabilities based on the combination of a soft trunk with a solid body. In our system, the rigid arm ensures positioning and orientation, mimicking the role of the elephant’s head, while the soft manipulator reproduces trunk motion primitives of bending and twisting under proper actuation patterns. This synergy replicates 9 distinct elephant grasping strategies reported in the literature, accommodating objects of varying shapes and sizes. The synergistic interaction between the rigid and soft components of the system minimizes the control complexity while maintaining a high degree of adaptability. Copyright © 2025, The Authors. All rights reserved.

关键词： Robotic arms

来源：评论

学校读者我要写书评

暂无评论

Rewiring of SINE-MIR enhancer topology and Esrrb modulation in expanded and naive pluripotency

引用

Genome Biology 2025年第1期26卷 1-32页

作者： Cipta, Nadia Omega Zeng, Yingying Wong, Ka Wai Zheng, Zi Hao Yi, Yao Warrier, Tushar Teo, Jian Zhou Teo, Jia Hao Jackie Kok, Yee Jiun Bi, Xuezhi Taneja, Reshma Ong, Derrick Sek Tong Xu, Jian Ginhoux, Florent Li, Hu Liou, Yih-Cherng Loh, Yuin-Han Epigenetics and Cell Fates Laboratory Cell Fate Engineering and Therapeutics Laboratory Institute of Molecular and Cell Biology (IMCB) Agency for Science Technology and Research (A*STAR) 61 Biopolis Drive Proteos Singapore 138673 Singapore Department of Biological Sciences Faculty of Science National University of Singapore 14 Science Drive 4 Singapore 117543 Singapore School of Biological Sciences Nanyang Technological University Singapore Singapore Department of Physiology NUS Yong Loo Lin School of Medicine 2 Medical Drive MD9 Singapore Singapore Proteomics Group Agency for Science Technology and Research (A*STAR) Bioprocessing Technology Institute (BTI) Singapore 138668 Singapore NUS Center for Cancer Research Yong Loo Lin School of Medicine National University of Singapore Singapore 117597 Singapore Department of Pathology Center of Excellence for Leukemia Studies St. Jude Children’s Research Hospital Memphis 38105 TN United States Singapore Immunology Network (SIgN) Agency for Science Technology and Research (A*STAR) Singapore Singapore Department of Immunology and Microbiology Shanghai Institute of Immunology Shanghai Jiao Tong University School of Medicine Shanghai China INSERM U1015 Paris Saclay University Gustave Roussy Cancer Campus Villejuif France Translational Immunology Institute SingHealth Duke-NUS Academic Medical Centre Singapore Singapore Department of Microbiology and Immunology Yong Loo Lin School of Medicine National University of Singapore Singapore Singapore Department of Molecular Pharmacology & Experimental Therapeutics Center for Individualized Medicine Mayo Clinic Rochester 55905 MN United States NUS Graduate School’s Integrative Sciences and Engineering Programme National University of Singapore 28 Medical Drive Singapore Singapore

Background: The interplay between 3D genomic structure and transposable elements (TE) in regulating cell state-specific gene expression program is largely unknown. Here, we explore the utilization of TE-derived enhancers in naïve and expanded pluripotent states by integrative analysis of genome-wide Hi-C-defined enhancer interactions, H3K27ac HiChIP profiling and CRISPR-guided TE proteomics landscape. Results: We find that short interspersed nuclear elements (SINEs) are the more involved TEs in the active chromatin and 3D genome architecture. In particular, mammalian-wide interspersed repeat (MIR), a SINE family member, is highly associated with naïve-specific genomic interactions compared to the expanded state. Primarily, in the naïve pluripotent state, MIR enhancer is co-opted by ESRRB for naïve-specific gene expression program. This ESRRB and MIR enhancer interaction is crucial for the formation of loops that build a network of enhancers and super-enhancers regulating pluripotency genes. We demonstrate that loss of a ESRRB-bound MIR enhancer impairs self-renewal. We also find that MIR is co-bound by structural protein complex, ESRRB-YY1, in the naïve pluripotent state. Conclusions: Altogether, our study highlights the topological regulation of ESRRB on MIR in the naïve potency state. © The Author(s) 2025.

关键词： 3D genome Mouse embryonic stem cells Pluripotency Transposable element

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：