3,577 results match your criteria: "University of Defense[Affiliation]"

Enhancing Rashba Spin-Splitting Strength by Orbital Hybridization.

ACS Nano

December 2024

Department of Materials Science and Engineering, National University of Singapore, Singapore117575, Singapore.

A Rashba spin-splitting state with spin-momentum locking enables the charge-spin interconversion known as the Rashba effect, induced by the interplay of inversion symmetry breaking (ISB) and spin-orbit coupling (SOC). Enhancing spin-splitting strength is promising to achieve high spin-orbit torque (SOT) efficiency for low-power-consumption spintronic devices. However, the energy scale of natural ISB at the interface is relatively small, leading to the weak Rashba effect.

View Article and Find Full Text PDF

PD-L2 act as an independent immune checkpoint in colorectal cancer beyond PD-L1.

Front Immunol

December 2024

Department of Oncology, The Second Xiangya Hospital, Central South University, Changsha, Hunan, China.

Introduction: Immunotherapy, especially immune checkpoint blockade (ICB), holds promise as a therapeutic strategy in colorectal cancer (CRC) by harnessing the patient's immune system to target malignant cells. Particularly, the PD-1/PD-L1 axis is widely recognized for its critical role in tumor microenvironment immunosuppression. Antibodies targeting PD-1 or PD-L1 have shown sustained efficacy against various cancers, including CRC.

View Article and Find Full Text PDF

Traditional multimodal contrastive learning brings text and its corresponding image closer together as a positive pair, where the text typically consists of fixed sentence structures or specific descriptive statements, and the image features are generally global features (with some fine-grained work using local features). Similar to unimodal self-supervised contrastive learning, this approach can be seen as enforcing a strict identity constraint in a multimodal context. However, due to the inherent complexity of remote sensing images, which cannot be easily described in a single sentence, and the fact that remote sensing images contain rich ancillary information beyond just object features, this strict identity constraint may be insufficient.

View Article and Find Full Text PDF

With the growing number of electronic devices loaded and increasing influence from electromagnetic interference, large-scale systems or platforms are confronted with increasingly severe electromagnetic compatibility challenges. Due to the vast size of these systems and the multitude of electronic devices they contain, standard laboratory environments are often inadequate for meeting test requirements. This paper reviews the state-of-art in the area of field measurement techniques related to the checking of electromagnetic compatibility, and the key technologies of electromagnetic interference filtering and wide-bandwidth, large-dynamic, and rapidly transient signal extraction in the measurement field are analyzed.

View Article and Find Full Text PDF

Design of a Deep Learning-Based Metalens Color Router for RGB-NIR Sensing.

Nanomaterials (Basel)

December 2024

State Key Laboratory of Pulsed Power Laser Technology, College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China.

Metalens can achieve arbitrary light modulation by controlling the amplitude, phase, and polarization of the incident waves and have been applied across various fields. This paper presents a color router designed based on metalens, capable of effectively separating spectra from visible light to near-infrared light. Traditional design methods for meta-lenses require extensive simulations, making them time-consuming.

View Article and Find Full Text PDF

A green chemical shear-thickening polishing (GC-STP) method was studied to improve the surface precision and processing efficiency of monocrystalline silicon. A novel green shear-thickening polishing slurry composed of silica nanoparticles, alumina abrasive, sorbitol, plant ash, polyethylene glycol, and deionized water was formulated. The monocrystalline silicon was roughly ground using a diamond polishing slurry and then the GC-STP process.

View Article and Find Full Text PDF

The promoting effect of Balanus albicostatus cement protein 19k (Balcp19k) on wound healing by regulating fibroblast migration and relieving early-stage inflammation responses.

Int J Biol Macromol

December 2024

Department of Vascular Surgery, The Second Xiangya Hospital, Central South University, Changsha, China; The Institute of Vascular Diseases, Central South University, Changsha, China. Electronic address:

The high incidence of skin wounds (e.g., burns, diabetic ulcers, venous ulcers, pressure sores, and radiation injuries) remains a significant challenge in clinical settings.

View Article and Find Full Text PDF

BIRDNN: Behavior-Imitation Based Repair for Deep Neural Networks.

Neural Netw

December 2024

College of Computer Science and Technology, National University of Defense Technology, Changsha, 410000, Hunan, China.

The increasing utilization of deep neural networks (DNNs) in safety-critical systems has raised concerns about their potential to exhibit undesirable behaviors. Consequently, DNN repair/patching arises in response to the times, and it aims to eliminate unexpected predictions generated by flawed DNNs. However, existing repair methods, both retraining- and fine-tuning-based, primarily focus on high-level abstract interpretations or inferences of state spaces, often neglecting the outputs of underlying neurons.

View Article and Find Full Text PDF

Generative named entity recognition framework for Chinese legal domain.

PeerJ Comput Sci

November 2024

School of Computer Science and Engineering, Central South University, Changsha, Hunan, China.

Named entity recognition (NER) is a crucial task in natural language processing, particularly challenging in the legal domain due to the intricate and lengthy nature of legal entities. Existing methods often struggle with accurately identifying entity boundaries and types in legal texts. To address these challenges, we propose a novel sequence-to-sequence framework designed specifically for the legal domain.

View Article and Find Full Text PDF

Photoinduced ultrafast multielectron transfer and long-lived charge-accumulated state in a fullerene-indacenodithiophene dumbbell triad.

Proc Natl Acad Sci U S A

December 2024

Beijing National Laboratory for Molecular Sciences, Key Laboratory of Molecular Nanostructure and Nanotechnology, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China.

Article Synopsis
  • Researchers developed a unique dumbbell-shaped molecule that facilitates ultrafast multielectron transfer (m-ET), which is essential for improving light-energy conversion.
  • The m-ET in this system occurs 23 times faster than the more common single-electron transfer (s-ET) and has a significantly longer product lifetime (10 μs), which enhances its photocatalytic capabilities.
  • This study showcases a new approach to achieving efficient m-ET in single molecules, offering potential advancements for applications in photocatalysis and molecular devices.
View Article and Find Full Text PDF
Article Synopsis
  • The demand for advanced camouflage technology in military warfare is rising, particularly for materials that have multispectral compatibility and adaptive capabilities.
  • Achieving simultaneous multispectral functionality in one device is difficult due to the limitations in how electromagnetic waves respond across different wavelengths.
  • The integration of coding metamaterials with infrared electrochromic devices leads to a new type of metadevice that allows for adjustable infrared emissivity and significant microwave scattering reduction, presenting a promising approach for future multispectral camouflage solutions.
View Article and Find Full Text PDF

Traditionally, Rayleigh scattering is thought to only impact fiber sensing system performance when the leading fiber is over 10 km long. However, this report illustrates theoretically and experimentally that Rayleigh scattering cannot be ignored in fiber optic interferometric sensor (FOIS) even with several hundred-meter common leading fiber because of the interaction of Rayleigh backward scattering (RBS) and returning interference signal. Herein, a conceptual framework is developed to elucidate the interaction between RBS and FOIS interference, revealing that, beyond laser monochromacity, the self-correction characteristic of laser pulses also influences coherent superposition.

View Article and Find Full Text PDF
Article Synopsis
  • The paper introduces a gate driver designed for IGBT fast switching in compact pulsed power systems, targeting applications in electromagnetic launch and food processing.
  • It features a planar transformer with a unique half-turn winding that reduces leakage inductance by 31.1%, which improves the rise time of the driving voltage pulse.
  • Testing showed that the gate driver can achieve a current change rate (di/dt) of about 10.10 A/ns with a gate voltage of 50 V and a charging time of 88 ns, confirming its effectiveness for rapid IGBT switching.
View Article and Find Full Text PDF

Vacancy-Activated Surface Reconstruction of Perovskite Nanofibers for Efficient Lattice Oxygen Evolution.

ACS Appl Mater Interfaces

December 2024

School of Minerals Processing and Bioengineering, Central South University, Changsha, Hunan 410083, China.

Article Synopsis
  • The study focuses on enhancing the oxygen evolution reaction (OER) in perovskites by inducing surface reconstruction through trace Ce-doped LaCeNiFeO nanofibers (LCNF-NFs), which improves their catalytic activity.* -
  • High oxygen vacancies in the LCNF-NFs lower the reconstruction potential and increase electrolyte access, leading to significant generation of self-reconstructed electroactive Ni/FeO(OH) species on the surface.* -
  • The restructured LCNF-NFs demonstrated superior performance with a lower Tafel slope of 50.12 mV dec and a reduced overpotential of 342.3 mV, making them highly efficient compared to conventional electrocatalysts.*
View Article and Find Full Text PDF

The positive hemodynamic response (PHR) during stimulation often co-occurs with a strong, sustained negative hemodynamic response (NHR). However, the characteristics and neurophysiological mechanisms of the NHR, especially in regions distal to the PHR, remain incompletely understood. Using intrinsic optical imaging (OI) and two-photon imaging, we observed that forelimb electrical stimulation evoked strong PHR signals in the forelimb region of the primary somatosensory cortex (S1FL).

View Article and Find Full Text PDF

The analysis of connection strengths and distances in the mobility network is pivotal for delineating critical pathways, particularly in the context of epidemic propagation. Local connections that link proximate districts typically exhibit strong weights. However, ties that bridge distant regions with high levels of interaction intensity, termed strong long (SL) ties, warrant increased scrutiny due to their potential to foster satellite epidemic clusters and extend the duration of pandemics.

View Article and Find Full Text PDF

The detection of bolt loosening in key components of aircraft engines faces problems such as complex and difficult-to-establish bolt loosening mechanism models, difficulty in identifying early loosening, and difficulty in extracting signal features with nonlinear and non-stationary characteristics. Therefore, the automated structural bolt micro looseness monitoring method using deep learning was proposed. Specifically, the addition of batch normalization methods enables the established Batch Normalized Stacked Autoencoders (BNSAEs) model to converge quickly and effectively, making the model easy to build and effective.

View Article and Find Full Text PDF

A photonic lantern is a low-loss device that connects a single multimode waveguide to multiple single-mode waveguides and can enhance the beam quality of a fiber laser by adaptively controlling the optical parameters (amplitude, phase, polarization) at the input. In this work, we combined the gains and losses of individual modes within the fiber amplifier and introduced a mode content parameter at the amplifier's output as an evaluation function to simulate mode control effects. Mode competition within the gain fiber can degrade the control effect of the fundamental mode and lead to it taking a longer time for the control to converge.

View Article and Find Full Text PDF
Article Synopsis
  • Genomic analysis is essential for understanding species evolution and conservation, but the complete chloroplast genome of a rare Chinese medicinal plant has not been reported, hindering its conservation efforts.
  • In this study, researchers sequenced the first complete chloroplast genome using Illumina technology, revealing a typical structure with 123 genes and significant genetic variations.
  • The findings suggest evolutionary relationships among related species and highlight the genome's potential for guiding further biological studies and conservation strategies.
View Article and Find Full Text PDF

Background: Lour., a genus in the subfamily Cercidoideae of the family Leguminosae, is characterized by woody liana habit, tendrils, and distinctive bilobate or bifoliolate leaves. The genus holds important medicinal value and constitutes a complex group characterized by morphological diversity and unstable taxonomic boundaries.

View Article and Find Full Text PDF

Toward Transparent and Controllable Quantum Generative Models.

Entropy (Basel)

November 2024

Department of Intelligent Data Science, College of Computer Science and Technology, National University of Defense Technology, Changsha 410073, China.

Quantum generative models have shown promise in fields such as quantum chemistry, materials science, and optimization. However, their practical utility is hindered by a significant challenge: the lack of interpretability. In this work, we introduce to enhance both the interpretability and controllability of quantum generative models.

View Article and Find Full Text PDF

Mapping Data to Concepts: Enhancing Quantum Neural Network Transparency with Concept-Driven Quantum Neural Networks.

Entropy (Basel)

October 2024

Department of Intelligent Data Science, College of Computer Science and Technology, National University of Defense Technology, Changsha 410073, China.

We introduce the concept-driven quantum neural network (CD-QNN), an innovative architecture designed to enhance the interpretability of quantum neural networks (QNNs). CD-QNN merges the representational capabilities of QNNs with the transparency of self-explanatory models by mapping input data into a human-understandable concept space and making decisions based on these concepts. The algorithmic design of CD-QNN is comprehensively analyzed, detailing the roles of the concept generator, feature extractor, and feature integrator in improving and balancing model expressivity and interpretability.

View Article and Find Full Text PDF

Image Forensics in the Encrypted Domain.

Entropy (Basel)

October 2024

College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China.

Encryption techniques used by forgers have thrown out a big possible challenge to forensics. Most traditional forensic tools will fail to detect the forged multimedia, which has been encrypted. Thus, image forensics in the encrypted domain (IFED) is significant.

View Article and Find Full Text PDF

The transition in medical image segmentation from fine-grained to coarse-grained annotation methods, notably scribble annotation, offers a practical and efficient preparation for deep learning applications. However, these methods often compromise segmentation precision and result in irregular contours. This study targets the enhancement of scribble-supervised segmentation to match the accuracy of fine-grained annotation.

View Article and Find Full Text PDF

CoSD: Balancing behavioral consistency and diversity in unsupervised skill discovery.

Neural Netw

February 2025

School of Computer Science and Technology, Soochow University, Suzhou, 215006, China. Electronic address:

In hierarchical reinforcement learning, unsupervised skill discovery holds promise for overcoming the challenge of sparse rewards commonly encountered in traditional reinforcement learning. Although previous unsupervised skill discovery methods excelled at maximizing intrinsic rewards, they often overly prioritized skill diversity. Unrestrained pursuit of diversity leads skills to concentrate attention on unexplored domains, overlooking the internal consistency of skills themselves, resulting in the state visit distribution of individual skills lacking concentration.

View Article and Find Full Text PDF