NSTL国家科技图书文献中心

1. Expressive Speech Synthesis Enhancement with Conditional Embeddings NSTL国家科技图书文献中心

Fanfan Yan | Maoyu Zhang ... - 《PRICAI 2024,Part IV》 - Pacific Rim International Conference on Artificial Intelligence - 2025, - 359～368 - 共10页

摘要： emotion prediction module, a prosody prediction module | predict emotion* labels based on the input text, or* | In recent years, emotional speech synthesis | techniques have attracted considerable interest because of | their wide-ranging potential applications. However

关键词： Speech synthesis | Expressive speech | Prosodic features | Emotion embedding

2. Multiplex graph aggregation and feature refinement for unsupervised incomplete multimodal emotion recognition NSTL国家科技图书文献中心

Deng, Yuanyue | Bian, Jintang ... - 《Information Fusion》 - 2025,114 - 共15页

摘要：Multimodal Emotion* Recognition (MER) involves* | : Completion, Aggregation, Refinement, and Embedding | operations on the embedding* features to obtain the fused* | separable and conducive to emotion* recognition* | . Experimental results on multimodal emotion* recognition*

关键词： Incomplete multimodal learning | Emotion recognition | Multiplex graph aggregation | Contrastive learning

3. Combining Graph NN and LLM for Improved Text-Based Emotion Recognition NSTL国家科技图书文献中心

Xinhao Zou | Konstantin Markov - 《Artificial Intelligence: Methodology,Systems,and Applications》 - International Conference on Artificial Intelligence: Methodology,Systems,and Applications - 2025, - 143～154 - 共12页

摘要：Text-based emotion* analysis, an important task* | in various emotion* recognition studies. They have* | to combine GNNs with LLMs for the emotion | embedding. For experiments and system evaluation, we used | that combining GNN and LLM improves the emotion

关键词： Graph neural network | Large language model | Emotion analysis and recognition

4. LLDif: Diffusion Models for Low-Light Facial Expression Recognition NSTL国家科技图书文献中心

Zhifeng Wang | Kaihao Zhang ... - 《Pattern Recognition,Part XIII》 - International Conference on Pattern Recognition - 2025, - 386～401 - 共16页

摘要： significantly reduce the accuracy of emotion* recognition* | -CLIP), an embedding* prior network (PNET), and a* | generating a joint embedding* prior distribution (EPD) to* | This paper introduces LLDif, a novel diffusion | -based facial expression recognition (FER) framework

关键词： Low-Light | Emotion recognition | Diffusion model

5. CoMaSa:Context Multi-aware Self-attention for emotional response generation NSTL国家科技图书文献中心

Zheng X. | Du Y.... - 《Neurocomputing》 - 2025,611(Jan.1) - 1.1～1.15 - 共15页

摘要： generating utterances with appropriate contextual emotion | dialogue context and the corresponding context emotion | self-attention emotion* dialogue generation model that* | embedding vectors, and enrich the dialogue's embedding | establish an emotion-aware decoder based on multi-task

关键词： Attention | Contextual emotion utilization | Emotional dialogue generation | External knowledge

6. Improving Chinese Emotion Classification Based on Bilingual Feature Fusion NSTL国家科技图书文献中心

Haocheng Lan | Jie Ou ... - 《Pattern Recognition,Part XXXI》 - International Conference on Pattern Recognition - 2025, - 139～153 - 共15页

摘要： challenge to current emotion* classification methods. To* | ) and Embedding* Fine-tuning (LEF) to reduce the* | emotion classification in the context of mixed-language | The growing popularity of Chinese social media | platforms such as Sina Weibo has created a large number of

关键词： Emotion classification | Bilingual feature fusion | Bilingual cross attention mechanism

7. Large Multimodal Models Thrive with Little Data for Image Emotion Prediction NSTL国家科技图书文献中心

Peng He | Mohamed Hussein ... - 《Pattern Recognition,Part I》 - International Conference on Pattern Recognition - 2025, - 298～313 - 共16页

摘要： methods for image emotion* prediction rely on neural* | networks to learn an image-to-emotion* mapping. Such* | LMM's output embedding* to a classification output. On* | Images can convey rich semantics and induce a | wide range of emotions in viewers. However

关键词： Emotion prediction | Vision | Large multimodal model

8. ELEMO: Elements Focused Emotion Recognition for Sticker Images NSTL国家科技图书文献中心

Min Luo | Boda Lin ... - 《Pattern Recognition and Computer Vision,Part V》 - Chinese Conference on Pattern Recognition and Computer Vision - 2025, - 231～245 - 共15页

摘要： online chats. Existing efforts in sticker emotion | leverage overlapping patch embedding* to preserve the* | Sticker images have become widely popular as | potent tools for emotional expression in contemporary | recognition typically integrate both global and local

关键词： Sticker emotion recognition | Local information continuity | Text-image consistency

9. Mentalization-based interventions in schools for enhancing socio-emotional competencies and positive behaviour: a systematic review NSTL国家科技图书文献中心

Gali,Chelouche-Dwek | Peter,Fonagy - 《European child & adolescent psychiatry》 - 2025,34(4) - 1295～1315 - 共21页

摘要： aspects of mentalizing, such as emotion-understanding | abilities, emotion* regulation, and mental health outcomes* | . Embedding* MBIs within school curriculums and evaluating* | Abstract Mentalization-based interventions | (MBIs) have been increasingly applied in school

关键词： Mentalization | School-based intervention | Socio-emotional development | Children | Adolescents | Systematic review

10. REDAffectiveLM : leveraging affect enriched embedding and transformer-based neural language model for readers' emotion detection NSTL国家科技图书文献中心

Kadan, Anoop | Deepak, P.... - 《Knowledge and information systems》 - 2024,66(12) - 7495～7525 - 共31页

摘要：, emotion* expressed by the writer and emotion elicited* | approach for readers' emotion* detection from short-text* | considerably enhance readers' emotion* detection. Since the* | ' emotion* detection isn't well explored, we conduct a* | conventional semantic embedding, affect enriched embedding

摘要： emotion prediction module, a prosody prediction module | predict emotion* labels based on the input text, or* | In recent years, emotional speech synthesis | techniques have attracted considerable interest because of | their wide-ranging potential applications. However

关键词： Speech synthesis | Expressive speech | Prosodic features | Emotion embedding

摘要：Multimodal Emotion* Recognition (MER) involves* | : Completion, Aggregation, Refinement, and Embedding | operations on the embedding* features to obtain the fused* | separable and conducive to emotion* recognition* | . Experimental results on multimodal emotion* recognition*

关键词： Incomplete multimodal learning | Emotion recognition | Multiplex graph aggregation | Contrastive learning

摘要：Text-based emotion* analysis, an important task* | in various emotion* recognition studies. They have* | to combine GNNs with LLMs for the emotion | embedding. For experiments and system evaluation, we used | that combining GNN and LLM improves the emotion

关键词： Graph neural network | Large language model | Emotion analysis and recognition

摘要： significantly reduce the accuracy of emotion* recognition* | -CLIP), an embedding* prior network (PNET), and a* | generating a joint embedding* prior distribution (EPD) to* | This paper introduces LLDif, a novel diffusion | -based facial expression recognition (FER) framework

关键词： Low-Light | Emotion recognition | Diffusion model

摘要： generating utterances with appropriate contextual emotion | dialogue context and the corresponding context emotion | self-attention emotion* dialogue generation model that* | embedding vectors, and enrich the dialogue's embedding | establish an emotion-aware decoder based on multi-task

关键词： Attention | Contextual emotion utilization | Emotional dialogue generation | External knowledge

摘要： challenge to current emotion* classification methods. To* | ) and Embedding* Fine-tuning (LEF) to reduce the* | emotion classification in the context of mixed-language | The growing popularity of Chinese social media | platforms such as Sina Weibo has created a large number of

关键词： Emotion classification | Bilingual feature fusion | Bilingual cross attention mechanism

摘要： methods for image emotion* prediction rely on neural* | networks to learn an image-to-emotion* mapping. Such* | LMM's output embedding* to a classification output. On* | Images can convey rich semantics and induce a | wide range of emotions in viewers. However

关键词： Emotion prediction | Vision | Large multimodal model

摘要： online chats. Existing efforts in sticker emotion | leverage overlapping patch embedding* to preserve the* | Sticker images have become widely popular as | potent tools for emotional expression in contemporary | recognition typically integrate both global and local

关键词： Sticker emotion recognition | Local information continuity | Text-image consistency

摘要： aspects of mentalizing, such as emotion-understanding | abilities, emotion* regulation, and mental health outcomes* | . Embedding* MBIs within school curriculums and evaluating* | Abstract Mentalization-based interventions | (MBIs) have been increasingly applied in school

关键词： Mentalization | School-based intervention | Socio-emotional development | Children | Adolescents | Systematic review

关键词： Readers' emotion detection | Textual emotion detection | Affective computing | Affect enriched embedding | Language model | Deep learning

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： emotion prediction module, a prosody prediction module | predict emotion labels based on the input text, or | In recent years, emotional speech synthesis | techniques have attracted considerable interest because of | their wide-ranging potential applications. However

关键词： Speech synthesis | Expressive speech | Prosodic features | Emotion embedding

摘要：Multimodal Emotion Recognition (MER) involves | : Completion, Aggregation, Refinement, and Embedding | operations on the embedding features to obtain the fused | separable and conducive to emotion recognition | . Experimental results on multimodal emotion recognition

关键词： Incomplete multimodal learning | Emotion recognition | Multiplex graph aggregation | Contrastive learning

摘要：Text-based emotion analysis, an important task | in various emotion recognition studies. They have | to combine GNNs with LLMs for the emotion | embedding. For experiments and system evaluation, we used | that combining GNN and LLM improves the emotion

关键词： Graph neural network | Large language model | Emotion analysis and recognition

摘要： significantly reduce the accuracy of emotion recognition | -CLIP), an embedding prior network (PNET), and a | generating a joint embedding prior distribution (EPD) to | This paper introduces LLDif, a novel diffusion | -based facial expression recognition (FER) framework

关键词： Low-Light | Emotion recognition | Diffusion model

摘要： generating utterances with appropriate contextual emotion | dialogue context and the corresponding context emotion | self-attention emotion dialogue generation model that | embedding vectors, and enrich the dialogue's embedding | establish an emotion-aware decoder based on multi-task

关键词： Attention | Contextual emotion utilization | Emotional dialogue generation | External knowledge

摘要： challenge to current emotion classification methods. To | ) and Embedding Fine-tuning (LEF) to reduce the | emotion classification in the context of mixed-language | The growing popularity of Chinese social media | platforms such as Sina Weibo has created a large number of

关键词： Emotion classification | Bilingual feature fusion | Bilingual cross attention mechanism

摘要： methods for image emotion prediction rely on neural | networks to learn an image-to-emotion mapping. Such | LMM's output embedding to a classification output. On | Images can convey rich semantics and induce a | wide range of emotions in viewers. However

关键词： Emotion prediction | Vision | Large multimodal model

摘要： online chats. Existing efforts in sticker emotion | leverage overlapping patch embedding to preserve the | Sticker images have become widely popular as | potent tools for emotional expression in contemporary | recognition typically integrate both global and local

关键词： Sticker emotion recognition | Local information continuity | Text-image consistency

摘要： aspects of mentalizing, such as emotion-understanding | abilities, emotion regulation, and mental health outcomes | . Embedding MBIs within school curriculums and evaluating | Abstract Mentalization-based interventions | (MBIs) have been increasingly applied in school

关键词： Mentalization | School-based intervention | Socio-emotional development | Children | Adolescents | Systematic review

摘要：, emotion expressed by the writer and emotion elicited | approach for readers' emotion detection from short-text | considerably enhance readers' emotion detection. Since the | ' emotion detection isn't well explored, we conduct a | conventional semantic embedding, affect enriched embedding

关键词： Readers' emotion detection | Textual emotion detection | Affective computing | Affect enriched embedding | Language model | Deep learning

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： emotion prediction module, a prosody prediction module | predict emotion* labels based on the input text, or* | In recent years, emotional speech synthesis | techniques have attracted considerable interest because of | their wide-ranging potential applications. However

摘要：Multimodal Emotion* Recognition (MER) involves* | : Completion, Aggregation, Refinement, and Embedding | operations on the embedding* features to obtain the fused* | separable and conducive to emotion* recognition* | . Experimental results on multimodal emotion* recognition*

摘要：Text-based emotion* analysis, an important task* | in various emotion* recognition studies. They have* | to combine GNNs with LLMs for the emotion | embedding. For experiments and system evaluation, we used | that combining GNN and LLM improves the emotion

摘要： significantly reduce the accuracy of emotion* recognition* | -CLIP), an embedding* prior network (PNET), and a* | generating a joint embedding* prior distribution (EPD) to* | This paper introduces LLDif, a novel diffusion | -based facial expression recognition (FER) framework

摘要： generating utterances with appropriate contextual emotion | dialogue context and the corresponding context emotion | self-attention emotion* dialogue generation model that* | embedding vectors, and enrich the dialogue's embedding | establish an emotion-aware decoder based on multi-task

摘要： challenge to current emotion* classification methods. To* | ) and Embedding* Fine-tuning (LEF) to reduce the* | emotion classification in the context of mixed-language | The growing popularity of Chinese social media | platforms such as Sina Weibo has created a large number of

摘要： methods for image emotion* prediction rely on neural* | networks to learn an image-to-emotion* mapping. Such* | LMM's output embedding* to a classification output. On* | Images can convey rich semantics and induce a | wide range of emotions in viewers. However

摘要： online chats. Existing efforts in sticker emotion | leverage overlapping patch embedding* to preserve the* | Sticker images have become widely popular as | potent tools for emotional expression in contemporary | recognition typically integrate both global and local

摘要： aspects of mentalizing, such as emotion-understanding | abilities, emotion* regulation, and mental health outcomes* | . Embedding* MBIs within school curriculums and evaluating* | Abstract Mentalization-based interventions | (MBIs) have been increasingly applied in school