With the growing demand formore comprehensive and nuanced sentiment understanding,Multimodal Sentiment Analysis(MSA)has gained significant traction in recent years and continues to attract widespread attention in the ...With the growing demand formore comprehensive and nuanced sentiment understanding,Multimodal Sentiment Analysis(MSA)has gained significant traction in recent years and continues to attract widespread attention in the academic community.Despite notable advances,existing approaches still face critical challenges in both information modeling and modality fusion.On one hand,many current methods rely heavily on encoders to extract global features from each modality,which limits their ability to capture latent fine-grained emotional cues within modalities.On the other hand,prevailing fusion strategies often lack mechanisms to model semantic discrepancies across modalities and to adaptively regulate modality interactions.To address these limitations,we propose a novel framework for MSA,termed Multi-Granularity Guided Fusion(MGGF).The proposed framework consists of three core components:(i)Multi-Granularity Feature Extraction Module,which simultaneously captures both global and local emotional features within each modality,and integrates them to construct richer intra-modal representations;(ii)Cross-ModalGuidance Learning Module(CMGL),which introduces a cross-modal scoring mechanism to quantify the divergence and complementarity betweenmodalities.These scores are then used as guiding signals to enable the fusion strategy to adaptively respond to scenarios of modality agreement or conflict;(iii)Cross-Modal Fusion Module(CMF),which learns the semantic dependencies among modalities and facilitates deep-level emotional feature interaction,thereby enhancing sentiment prediction with complementary information.We evaluate MGGF on two benchmark datasets:MVSA-Single and MVSA-Multiple.Experimental results demonstrate that MGGF outperforms the current state-of-the-art model CLMLF on MVSA-Single by achieving a 2.32% improvement in F1 score.On MVSA-Multiple,it surpasses MGNNS with a 0.26% increase in accuracy.These results substantiate the effectiveness ofMGGFin addressing two major limitations of existing methods—insufficient intra-modal fine-grained sentiment modeling and inadequate cross-modal semantic fusion.展开更多
Under the background of new curriculum reform, how to improve students' autonomous learning ability is the core proposition of contemporary education reform. In the modern classroom, the application of the "g...Under the background of new curriculum reform, how to improve students' autonomous learning ability is the core proposition of contemporary education reform. In the modern classroom, the application of the "guided learning plan" has promoted the improvement of the traditional teaching model and gradually become an important support for students' autonomous learning. The information technology classroom in senior high school is internally organized and logical, which is very suitable for the application of the guided learning plan and can promote students' deep cognition of information technology learning. Taking the high school information technology class as an example, this paper explores the effective ways to use the guided learning plan in information technology teaching, and puts forward personal opinions for the application of the guidance case in education and teaching.展开更多
Objective:To explore the application effect of constructing professional teaching staff in low-level training in operating room,so as to further optimize the teaching strength of operating room in our hospital and imp...Objective:To explore the application effect of constructing professional teaching staff in low-level training in operating room,so as to further optimize the teaching strength of operating room in our hospital and improve the training effect of junior nurses.Methods:Thirty-eight low-level nurses in the operating room of a third-class hospital in Yantai were selected for half a year's nurse training.With theoretical scores,overall nursing performance,nurses'self-awareness evaluation system and nurses'satisfaction with tutors as evaluation criteria,and based on the selection of high-quality teachers,various evaluation indexes before and after the training of low-level nurses in the operating room were compared and evaluated through the cultivation of practical teaching teachers'ability and the application of a series of teaching methods based on the change of c ompetence-based education(CBE)teaching mode,the application of guided learning interactive canadian education(BOPPPS)teaching model and p roblem-b ased l earning(PBL)teaching method.Results:After the training,the examination scores of low-level nurses were significantly improved(P<0.05),the teaching quality was highly recognized by low-level nurses,the quality of low-level nurses was improved,and patients'satisfaction with nurses was improved.Conclusion:It is of great significance to assist the construction of professional teachers in evidence-based medicine.Through the training of practical teaching teachers'ability and the application of a series of teaching methods based on the change of CBE teaching model,the application of BOPPPS teaching model and PBL teaching method,the training results of nurses have been significantly improved and improved,which is worthy of clinical reference and promotion.展开更多
Artificial intelligence has achieved remarkable success in materials science,accelerating novel material design.However,real-world material systems exhibit multiscale complexity—spanning composition,processing,struct...Artificial intelligence has achieved remarkable success in materials science,accelerating novel material design.However,real-world material systems exhibit multiscale complexity—spanning composition,processing,structure,and properties—posing significant challenges for modeling.While some approaches fuse multiscale features to improve prediction,important modalities such as microstructure are often missing due to high acquisition costs.Existing methods struggle with incomplete data and lack a framework to bridge multiscale material knowledge.To address this,we propose MatMCL,a structure-guided multimodal learning framework that jointly analyzes multiscale material information and enables robust property prediction with incomplete modalities.Using a selfconstructed multimodal dataset of electrospun nanofibers,we demonstrate that MatMCL improves mechanical property prediction without structural information,generates microstructures from processing parameters,and enables cross-modal retrieval.We further extend it via multi-stage learning and apply it to nanofiber-reinforced composite design.MatMCL uncovers processingstructure-property relationships,suggesting its promise as a generalizable approach for AI-driven material design.展开更多
Driven by advancements in artificial intelligence,end-to-end learning has become a key method for system optimization in various fields,including communications.However,applying learning algorithms such as backpropaga...Driven by advancements in artificial intelligence,end-to-end learning has become a key method for system optimization in various fields,including communications.However,applying learning algorithms such as backpropagation directly to communication systems is challenging due to their non-differentiable nature.Existing methods typically require developing a precise differentiable digital model of the physical system,which is computationally complex and can cause significant performance loss after deployment.In response,we propose a novel end-to-end learning framework called physics-guided learning.展开更多
基金supported in part by the National Key Research and Development Program of China under Grant 2022YFB3102904in part by the National Natural Science Foundation of China under Grant No.U23A20305 and No.62472440.
文摘With the growing demand formore comprehensive and nuanced sentiment understanding,Multimodal Sentiment Analysis(MSA)has gained significant traction in recent years and continues to attract widespread attention in the academic community.Despite notable advances,existing approaches still face critical challenges in both information modeling and modality fusion.On one hand,many current methods rely heavily on encoders to extract global features from each modality,which limits their ability to capture latent fine-grained emotional cues within modalities.On the other hand,prevailing fusion strategies often lack mechanisms to model semantic discrepancies across modalities and to adaptively regulate modality interactions.To address these limitations,we propose a novel framework for MSA,termed Multi-Granularity Guided Fusion(MGGF).The proposed framework consists of three core components:(i)Multi-Granularity Feature Extraction Module,which simultaneously captures both global and local emotional features within each modality,and integrates them to construct richer intra-modal representations;(ii)Cross-ModalGuidance Learning Module(CMGL),which introduces a cross-modal scoring mechanism to quantify the divergence and complementarity betweenmodalities.These scores are then used as guiding signals to enable the fusion strategy to adaptively respond to scenarios of modality agreement or conflict;(iii)Cross-Modal Fusion Module(CMF),which learns the semantic dependencies among modalities and facilitates deep-level emotional feature interaction,thereby enhancing sentiment prediction with complementary information.We evaluate MGGF on two benchmark datasets:MVSA-Single and MVSA-Multiple.Experimental results demonstrate that MGGF outperforms the current state-of-the-art model CLMLF on MVSA-Single by achieving a 2.32% improvement in F1 score.On MVSA-Multiple,it surpasses MGNNS with a 0.26% increase in accuracy.These results substantiate the effectiveness ofMGGFin addressing two major limitations of existing methods—insufficient intra-modal fine-grained sentiment modeling and inadequate cross-modal semantic fusion.
文摘Under the background of new curriculum reform, how to improve students' autonomous learning ability is the core proposition of contemporary education reform. In the modern classroom, the application of the "guided learning plan" has promoted the improvement of the traditional teaching model and gradually become an important support for students' autonomous learning. The information technology classroom in senior high school is internally organized and logical, which is very suitable for the application of the guided learning plan and can promote students' deep cognition of information technology learning. Taking the high school information technology class as an example, this paper explores the effective ways to use the guided learning plan in information technology teaching, and puts forward personal opinions for the application of the guidance case in education and teaching.
文摘Objective:To explore the application effect of constructing professional teaching staff in low-level training in operating room,so as to further optimize the teaching strength of operating room in our hospital and improve the training effect of junior nurses.Methods:Thirty-eight low-level nurses in the operating room of a third-class hospital in Yantai were selected for half a year's nurse training.With theoretical scores,overall nursing performance,nurses'self-awareness evaluation system and nurses'satisfaction with tutors as evaluation criteria,and based on the selection of high-quality teachers,various evaluation indexes before and after the training of low-level nurses in the operating room were compared and evaluated through the cultivation of practical teaching teachers'ability and the application of a series of teaching methods based on the change of c ompetence-based education(CBE)teaching mode,the application of guided learning interactive canadian education(BOPPPS)teaching model and p roblem-b ased l earning(PBL)teaching method.Results:After the training,the examination scores of low-level nurses were significantly improved(P<0.05),the teaching quality was highly recognized by low-level nurses,the quality of low-level nurses was improved,and patients'satisfaction with nurses was improved.Conclusion:It is of great significance to assist the construction of professional teachers in evidence-based medicine.Through the training of practical teaching teachers'ability and the application of a series of teaching methods based on the change of CBE teaching model,the application of BOPPPS teaching model and PBL teaching method,the training results of nurses have been significantly improved and improved,which is worthy of clinical reference and promotion.
基金supported by the National Key Research and Development Program of China(2022YFB3807300)Zhejiang Provincial Natural Science Foundation of China(LR25E030001)+2 种基金the Key Research and Development Project of Zhejiang Province(2024C03073)the financial support from the State Key Laboratory of Transvascular Implantation Devices(012024019)Transvascular Implantation Devices Research Institute China(TIDRIC)(KY012024007,KY012024009).
文摘Artificial intelligence has achieved remarkable success in materials science,accelerating novel material design.However,real-world material systems exhibit multiscale complexity—spanning composition,processing,structure,and properties—posing significant challenges for modeling.While some approaches fuse multiscale features to improve prediction,important modalities such as microstructure are often missing due to high acquisition costs.Existing methods struggle with incomplete data and lack a framework to bridge multiscale material knowledge.To address this,we propose MatMCL,a structure-guided multimodal learning framework that jointly analyzes multiscale material information and enables robust property prediction with incomplete modalities.Using a selfconstructed multimodal dataset of electrospun nanofibers,we demonstrate that MatMCL improves mechanical property prediction without structural information,generates microstructures from processing parameters,and enables cross-modal retrieval.We further extend it via multi-stage learning and apply it to nanofiber-reinforced composite design.MatMCL uncovers processingstructure-property relationships,suggesting its promise as a generalizable approach for AI-driven material design.
基金RGC(ECS 24203724,YCRG C4004-24Y,YCRG C1002-22Y)NSFC(62405258)+2 种基金Innovation and Technology Fund(ITS/237/22)Innovation and TechnologyFund(ITS/226/21FP)NSFC/RGC Joint Research Scheme(N_CUHK444/22).
文摘Driven by advancements in artificial intelligence,end-to-end learning has become a key method for system optimization in various fields,including communications.However,applying learning algorithms such as backpropagation directly to communication systems is challenging due to their non-differentiable nature.Existing methods typically require developing a precise differentiable digital model of the physical system,which is computationally complex and can cause significant performance loss after deployment.In response,we propose a novel end-to-end learning framework called physics-guided learning.
