In the past decade, Fake Base Stations (FBS) have been consistently employed by criminals to target mobile users through spam text messages. despite the introduction of several techniques to mitigate this problem, spa...
详细信息
dynamic metasurface antenna (dMA) array is a new type of array structure with low complexity and low cost, which is emerging as a promising technique for next-generation wireless networks. In this paper, a downlink mu...
详细信息
Website security detection is important for Internet security. Existing machine learning-based malicious UrL detection methods have a low accuracy and weak generalization ability. Thus, we proposed a new multi-feature...
详细信息
To correctly evaluate the safety of mechanical components, considering manufacturing inducedresidual stresses is essential. In this study, the risk analysis of fatigue crack initiation and propagation in cylindrical ...
详细信息
Proteomics is a new technology that has been widely applied in the field of life and health *** effectively addresses issues related to the impact of dietary structure on organs,tissues,and cells,as well as the change...
详细信息
Proteomics is a new technology that has been widely applied in the field of life and health *** effectively addresses issues related to the impact of dietary structure on organs,tissues,and cells,as well as the changes in proteins in various organs,tissues,and cells underdisease *** differential proteins identified through proteomics can serve as disease biomarkers and target proteins affecting health and can be used fordisease diagnosis and health *** this paper,the application of proteomics in the field of infl ammation in recent years was summarized,especially in the therapeutic target and mechanism of action,which opens up a new way for more effective prevention,diagnosis,and treatment of inflammation,and provides medical protection for human life and health.
To improve the accuracy of indirect tensile strength for a transversely isotropic rock in the Brazilian test, this study considered the three-dimensional (3d) deformation and the nonlinear stress–strain relationship....
详细信息
To improve the accuracy of indirect tensile strength for a transversely isotropic rock in the Brazilian test, this study considered the three-dimensional (3d) deformation and the nonlinear stress–strain relationship. A parametric study of a numerical Brazilian test was performed for a general range of elastic constants, revealing that the 3d modeling evaluated the indirect tensile strength up to 40% higher than the plane stress modeling. For the actual Asan gneiss, the 3d model evaluated the indirect tensile strength up to 10% higher and slightly enhanced the accuracy of deformation estimation compared with the plane stress model. The nonlinearity in stress–strain curve of Asan gneiss under uniaxial compression was then considered, such that the evaluated indirect tensile strength was affected by up to 10% and its anisotropy agreed well with the physical intuition. The estimation of deformation was significantly enhanced. The further validation on the nonlinear model is expected as future research.
during the inspection process of seamless steel tubes, the circumferential dynamic deviation of the multi-hub serpentine robot complicates the accurate estimation and control of its axial operating state. To address t...
详细信息
Overlay in upper layer with respect to lower layer need to be measured and controlled with sub-nanometer-level accuracy in semiconductor manufacturing process. However, as the metrology to device (MTd) deviation conti...
详细信息
Surface/interface engineering of a multimetallic nanostructure with diverse electrocatalytic properties fordirect liquid fuel cells is desirable yet ***,using visible light,a class of quaternary Pt_(1)Ag_(0.1)Bi_(0.1...
详细信息
Surface/interface engineering of a multimetallic nanostructure with diverse electrocatalytic properties fordirect liquid fuel cells is desirable yet ***,using visible light,a class of quaternary Pt_(1)Ag_(0.1)Bi_(0.16)Te_(0.29)ultrathin nanosheets is fabricated and used as high-performance anode electrocatalysts for formic acid-/alcohol-air fuel *** modified electronic structure of Pt,enhanced hydroxyl adsorption,and abundant exteriordefects afford Pt_(1)Ag_(0.1)Bi_(0.16)Te_(0.29)/C high intrinsic anodic electrocatalytic activity to boost the powerdensities of direct formic acid-/methanol-/ethanol-/ethylene glycol-/glycerol-air fuel cells,and the corresponding peak powerdensity of Pt_(1)Ag_(0.1)Bi_(0.16)Te_(0.29)/C is respectively 129.7,142.3,105.4,124.3,and 128.0 mW cm^(-2),considerably outperforming Pt/*** in situ Fourier transform infraredreflection spectroscopy reveals that formic acid oxidation on Pt_(1)Ag_(0.1)Bi_(0.16)Te_(0.29)/C occurs via a CO_(2)-free direct *** functional theory calculations show that the presence of Ag,Bi,and Te in Pt_(1)Ag_(0.1)Bi_(0.16)Te_(0.29)suppresses CO^(*)formation while optimizing dehydrogenation steps and synergistic effect and modified Pt effectively enhance H_(2)O dissociation to improve electrocatalytic *** synthesis strategy can be extended to 43 other types of ultrathin multimetallic nanosheets(from ternary to octonary nanosheets),and efficiently capture precious metals(i.e.,Pd,Pt,rh,ru,Au,and Ag)from different water sources.
Multi-modal Named Entity recognition (MNEr) aims to better identify meaningful textual entities by integrating information from images. Previous work has focused on extracting visual semantics at a fine-grained level,...
详细信息
Multi-modal Named Entity recognition (MNEr) aims to better identify meaningful textual entities by integrating information from images. Previous work has focused on extracting visual semantics at a fine-grained level, or obtaining entity related external knowledge from knowledge bases or Large Language Models (LLMs). However, these approaches ignore the poor semantic correlation between visual and textual modalities in MNErdatasets anddo not explore different multi-modal fusion approaches. In this paper, we present MMAVK, a multi-modal named entity recognition model with auxiliary visual knowledge and word-level fusion, which aims to leverage the Multi-modal Large Language Model (MLLM) as an implicit knowledge base. It also extracts vision-based auxiliary knowledge from the image for more accurate and effective recognition. Specifically, we propose vision-based auxiliary knowledge generation, which guides the MLLM to extract external knowledge exclusively derived from images to aid entity recognition by designing target-specific prompts, thus avoiding redundant recognition and cognitive confusion caused by the simultaneous processing of image-text pairs. Furthermore, we employ a word-level multi-modal fusion mechanism to fuse the extracted external knowledge with each word-embedding embedded from the transformer-based encoder. Extensive experimental results demonstrate that MMAVK outperforms or equals the state-of-the-art methods on the two classical MNErdatasets, even when the large models employed have significantly fewer parameters than other baselines.
暂无评论