Multimodal sentiment analysis aims to understand emotions from text,speech,and video data.However,current methods often overlook the dominant role of text and suffer from feature loss during integration.Given the vary...Multimodal sentiment analysis aims to understand emotions from text,speech,and video data.However,current methods often overlook the dominant role of text and suffer from feature loss during integration.Given the varying importance of each modality across different contexts,a central and pressing challenge in multimodal sentiment analysis lies in maximizing the use of rich intra-modal features while minimizing information loss during the fusion process.In response to these critical limitations,we propose a novel framework that integrates spatial position encoding and fusion embedding modules to address these issues.In our model,text is treated as the core modality,while speech and video features are selectively incorporated through a unique position-aware fusion process.The spatial position encoding strategy preserves the internal structural information of speech and visual modalities,enabling the model to capture localized intra-modal dependencies that are often overlooked.This design enhances the richness and discriminative power of the fused representation,enabling more accurate and context-aware sentiment prediction.Finally,we conduct comprehensive evaluations on two widely recognized standard datasets in the field—CMU-MOSI and CMU-MOSEI to validate the performance of the proposed model.The experimental results demonstrate that our model exhibits good performance and effectiveness for sentiment analysis tasks.展开更多
Optical vortex is a promising candidate for capacity scaling in next-generation optical communications.The generation of multi-vortex beams is of great importance for vortex-based optical communications.Traditional ap...Optical vortex is a promising candidate for capacity scaling in next-generation optical communications.The generation of multi-vortex beams is of great importance for vortex-based optical communications.Traditional approaches for generating multivortex beams are passive,unscalable and cumbersome.Here,we propose and demonstrate a multi-vortex laser,an active approach for creating multi-vortex beams directly at the source.By printing a specially-designed concentric-rings pattern on the cavity mirror,multi-vortex beams are generated directly from the laser.Spatially,the generated multi-vortex beams are decomposable and coaxial.Temporally,the multi-vortex beams can be simultaneously self-mode-locked,and each vortex component carries pulses with GHz-level repetition rate.Utilizing these distinct spatial-temporal characteristics,we demonstrate that the multi-vortex laser can be spatially and temporally encoded for data transmission,showing the potential of the developed multi-vortex laser in optical communications.The demonstrations may open up new perspectives for diverse applications enabled by the multi-vortex laser.展开更多
In many cases, high-resolution nuclear magnetic resonance （NMR） spectra are virtually impossible to obtain by con- ventional nuclear magnetic resonance methods because of inhomogeneity of magnetic field and inherent...In many cases, high-resolution nuclear magnetic resonance （NMR） spectra are virtually impossible to obtain by con- ventional nuclear magnetic resonance methods because of inhomogeneity of magnetic field and inherent heterogeneity of sample. Although conventional intramolecular zero-quantum coherence （ZQC） can be used to obtain high-resolution spectrum in inhomogeneous field, the acquisition takes rather long time. In this paper, a spatially encoded intramolecular ZQC technique is proposed to fast acquire high-resolution NMR spectrum in inhomogeneous field. For the first time, the gradient-driven decoding technique is employed to selectively acquire intramolecular ZQC signals. Theoretical analyses and experimental observations demonstrate that high-resolution NMR spectral information can be retrieved within several scans even when the field inhomogeneity is severe enough to erase most spectral information. This work provides a new way to enhance the acquisition efficiency of high-resolution intramolecular ZQC spectroscopy in inhomogeneous fields.展开更多
1 Introduction Due to the complexity of traffic scenarios,the motion of agents is influenced not only by road geometry and traffic rules but also by surrounding agents,making trajectory prediction for autonomous vehic...1 Introduction Due to the complexity of traffic scenarios,the motion of agents is influenced not only by road geometry and traffic rules but also by surrounding agents,making trajectory prediction for autonomous vehicles exceptionally challenging.The movement pattern of a single vehicle is typically influenced by nearby vehicles and its surrounding environmental information.Social psychologists have pointed out that individuals often imitate or follow other members of a group[1],using them as a reference for their behavior,which leads to the frequent occurrence of the herd effect in vehicle movement patterns[2].展开更多
Turbulence,a complex multi-scale phenomenon inherent in fluid flow systems,presents critical challenges and opportunities for understanding physical mechanisms across scientific and engineering domains.Although high-r...Turbulence,a complex multi-scale phenomenon inherent in fluid flow systems,presents critical challenges and opportunities for understanding physical mechanisms across scientific and engineering domains.Although high-resolution(HR)turbulence data remain indispensable for advancing both theoretical insights and engineering solutions,their acquisition is severely limited by prohibitively high computational costs.While deep learning architectures show transformative potential in reconstructing high-fidelity flow representations from sparse measurements,current methodologies suffer from two inherent constraints:strict reliance on perfectly paired training data and inability to perform multi-scale reconstruction within a unified framework.To address these challenges,we propose HADF,a hash-adaptive dynamic fusion implicit network for turbulence reconstruction.Specifically,we develop a low-resolution(LR)consistency loss that facilitates effective model training under conditions of missing paired data,eliminating the conventional requirement for fully matched LR and HR datasets.We further employ hash-adaptive spatial encoding and dynamic feature fusion to extract turbulence features,mapping them with implicit neural representations for reconstruction at arbitrary resolutions.Experimental results demonstrate that HADF achieves superior performance in global reconstruction accuracy and local physical properties compared to state-of-the-art models.It precisely recovers fine turbulence details for partially unpaired data conditions and diverse resolutions by training only once while maintaining robustness against noise.展开更多
基金supported by the Collaborative Tackling Project of the Yangtze River Delta SciTech Innovation Community(Nos.2024CSJGG01503,2024CSJGG01500)Guangxi Key Research and Development Program(No.AB24010317)Jiangxi Provincial Key Laboratory of Electronic Data Control and Forensics(Jiangxi Police College)(No.2025JXJYKFJJ002).
文摘Multimodal sentiment analysis aims to understand emotions from text,speech,and video data.However,current methods often overlook the dominant role of text and suffer from feature loss during integration.Given the varying importance of each modality across different contexts,a central and pressing challenge in multimodal sentiment analysis lies in maximizing the use of rich intra-modal features while minimizing information loss during the fusion process.In response to these critical limitations,we propose a novel framework that integrates spatial position encoding and fusion embedding modules to address these issues.In our model,text is treated as the core modality,while speech and video features are selectively incorporated through a unique position-aware fusion process.The spatial position encoding strategy preserves the internal structural information of speech and visual modalities,enabling the model to capture localized intra-modal dependencies that are often overlooked.This design enhances the richness and discriminative power of the fused representation,enabling more accurate and context-aware sentiment prediction.Finally,we conduct comprehensive evaluations on two widely recognized standard datasets in the field—CMU-MOSI and CMU-MOSEI to validate the performance of the proposed model.The experimental results demonstrate that our model exhibits good performance and effectiveness for sentiment analysis tasks.
基金supported by the National Natural Science Foundation of China Grant No.61675130,11774116,11721091,61490713,91850203,61761130082,11574001,the National Key R&D Program of China(2018YFB2200204,2018YFB1801803)the Royal Society-Newton Advanced Fellowship,the Natural Science Foundation of Hubei Province of China(2018CFA048)+2 种基金the Key R&D Program of Guangdong Province(2018B030325002)the Program for HUST Academic Frontier Youth Team(2016QYTD05)the Fundamental Research Funds for the Central Universities(2019kfyRCPY037).
文摘Optical vortex is a promising candidate for capacity scaling in next-generation optical communications.The generation of multi-vortex beams is of great importance for vortex-based optical communications.Traditional approaches for generating multivortex beams are passive,unscalable and cumbersome.Here,we propose and demonstrate a multi-vortex laser,an active approach for creating multi-vortex beams directly at the source.By printing a specially-designed concentric-rings pattern on the cavity mirror,multi-vortex beams are generated directly from the laser.Spatially,the generated multi-vortex beams are decomposable and coaxial.Temporally,the multi-vortex beams can be simultaneously self-mode-locked,and each vortex component carries pulses with GHz-level repetition rate.Utilizing these distinct spatial-temporal characteristics,we demonstrate that the multi-vortex laser can be spatially and temporally encoded for data transmission,showing the potential of the developed multi-vortex laser in optical communications.The demonstrations may open up new perspectives for diverse applications enabled by the multi-vortex laser.
基金supported by the National Natural Science Foundation of China(Grant Nos.11275161 and 11105114)
文摘In many cases, high-resolution nuclear magnetic resonance （NMR） spectra are virtually impossible to obtain by con- ventional nuclear magnetic resonance methods because of inhomogeneity of magnetic field and inherent heterogeneity of sample. Although conventional intramolecular zero-quantum coherence （ZQC） can be used to obtain high-resolution spectrum in inhomogeneous field, the acquisition takes rather long time. In this paper, a spatially encoded intramolecular ZQC technique is proposed to fast acquire high-resolution NMR spectrum in inhomogeneous field. For the first time, the gradient-driven decoding technique is employed to selectively acquire intramolecular ZQC signals. Theoretical analyses and experimental observations demonstrate that high-resolution NMR spectral information can be retrieved within several scans even when the field inhomogeneity is severe enough to erase most spectral information. This work provides a new way to enhance the acquisition efficiency of high-resolution intramolecular ZQC spectroscopy in inhomogeneous fields.
基金supported in part by the National Key R&D Program of China under Grant(2023YFF0612102)the Shandong Province Natural Science Foundation(ZR2024MF023)the Key Technology Research and Industrial Demonstration Projects in Qingdao City(23-7-2-qljh-4-gx,24-1-2-qljh-19-gx).
文摘1 Introduction Due to the complexity of traffic scenarios,the motion of agents is influenced not only by road geometry and traffic rules but also by surrounding agents,making trajectory prediction for autonomous vehicles exceptionally challenging.The movement pattern of a single vehicle is typically influenced by nearby vehicles and its surrounding environmental information.Social psychologists have pointed out that individuals often imitate or follow other members of a group[1],using them as a reference for their behavior,which leads to the frequent occurrence of the herd effect in vehicle movement patterns[2].
基金Project supported by the National Natural Science Foundation of China(No.12402349)the Natural Science Foundation of Hunan Province(No.2024JJ6468)+1 种基金the Youth Foundation of the National University of Defense Technology(No.ZK2023-11)the National Key Research and Development Program of China(No.2021YFB0300101)。
文摘Turbulence,a complex multi-scale phenomenon inherent in fluid flow systems,presents critical challenges and opportunities for understanding physical mechanisms across scientific and engineering domains.Although high-resolution(HR)turbulence data remain indispensable for advancing both theoretical insights and engineering solutions,their acquisition is severely limited by prohibitively high computational costs.While deep learning architectures show transformative potential in reconstructing high-fidelity flow representations from sparse measurements,current methodologies suffer from two inherent constraints:strict reliance on perfectly paired training data and inability to perform multi-scale reconstruction within a unified framework.To address these challenges,we propose HADF,a hash-adaptive dynamic fusion implicit network for turbulence reconstruction.Specifically,we develop a low-resolution(LR)consistency loss that facilitates effective model training under conditions of missing paired data,eliminating the conventional requirement for fully matched LR and HR datasets.We further employ hash-adaptive spatial encoding and dynamic feature fusion to extract turbulence features,mapping them with implicit neural representations for reconstruction at arbitrary resolutions.Experimental results demonstrate that HADF achieves superior performance in global reconstruction accuracy and local physical properties compared to state-of-the-art models.It precisely recovers fine turbulence details for partially unpaired data conditions and diverse resolutions by training only once while maintaining robustness against noise.
