Guardado en:
Detalles Bibliográficos
Autores principales: Li, Jie, Ding, Shifei, Guo, Lili, Li, Xuan
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2506.18716
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911019437129728
author Li, Jie
Ding, Shifei
Guo, Lili
Li, Xuan
author_facet Li, Jie
Ding, Shifei
Guo, Lili
Li, Xuan
contents Emotion Recognition in Conversation (ERC) aims to detect the emotions of individual utterances within a conversation. Generating efficient and modality-specific representations for each utterance remains a significant challenge. Previous studies have proposed various models to integrate features extracted using different modality-specific encoders. However, they neglect the varying contributions of modalities to this task and introduce high complexity by aligning modalities at the frame level. To address these challenges, we propose the Multi-modal Anchor Gated Transformer with Knowledge Distillation (MAGTKD) for the ERC task. Specifically, prompt learning is employed to enhance textual modality representations, while knowledge distillation is utilized to strengthen representations of weaker modalities. Furthermore, we introduce a multi-modal anchor gated transformer to effectively integrate utterance-level representations across modalities. Extensive experiments on the IEMOCAP and MELD datasets demonstrate the effectiveness of knowledge distillation in enhancing modality representations and achieve state-of-the-art performance in emotion recognition. Our code is available at: https://github.com/JieLi-dd/MAGTKD.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18716
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-modal Anchor Gated Transformer with Knowledge Distillation for Emotion Recognition in Conversation
Li, Jie
Ding, Shifei
Guo, Lili
Li, Xuan
Machine Learning
Computation and Language
Emotion Recognition in Conversation (ERC) aims to detect the emotions of individual utterances within a conversation. Generating efficient and modality-specific representations for each utterance remains a significant challenge. Previous studies have proposed various models to integrate features extracted using different modality-specific encoders. However, they neglect the varying contributions of modalities to this task and introduce high complexity by aligning modalities at the frame level. To address these challenges, we propose the Multi-modal Anchor Gated Transformer with Knowledge Distillation (MAGTKD) for the ERC task. Specifically, prompt learning is employed to enhance textual modality representations, while knowledge distillation is utilized to strengthen representations of weaker modalities. Furthermore, we introduce a multi-modal anchor gated transformer to effectively integrate utterance-level representations across modalities. Extensive experiments on the IEMOCAP and MELD datasets demonstrate the effectiveness of knowledge distillation in enhancing modality representations and achieve state-of-the-art performance in emotion recognition. Our code is available at: https://github.com/JieLi-dd/MAGTKD.
title Multi-modal Anchor Gated Transformer with Knowledge Distillation for Emotion Recognition in Conversation
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2506.18716