Long-Short Distance Graph Neural Networks and Improved Curriculum Learning for Emotion Recognition in Conversation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Xinran, Xu, Xiujuan, Qiao, Jiaqi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909913795526656
author Li, Xinran
Xu, Xiujuan
Qiao, Jiaqi
author_facet Li, Xinran
Xu, Xiujuan
Qiao, Jiaqi
contents Emotion Recognition in Conversation (ERC) is a practical and challenging task. This paper proposes a novel multimodal approach, the Long-Short Distance Graph Neural Network (LSDGNN). Based on the Directed Acyclic Graph (DAG), it constructs a long-distance graph neural network and a short-distance graph neural network to obtain multimodal features of distant and nearby utterances, respectively. To ensure that long- and short-distance features are as distinct as possible in representation while enabling mutual influence between the two modules, we employ a Differential Regularizer and incorporate a BiAffine Module to facilitate feature interaction. In addition, we propose an Improved Curriculum Learning (ICL) to address the challenge of data imbalance. By computing the similarity between different emotions to emphasize the shifts in similar emotions, we design a "weighted emotional shift" metric and develop a difficulty measurer, enabling a training process that prioritizes learning easy samples before harder ones. Experimental results on the IEMOCAP and MELD datasets demonstrate that our model outperforms existing benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2507_15205
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Long-Short Distance Graph Neural Networks and Improved Curriculum Learning for Emotion Recognition in Conversation
Li, Xinran
Xu, Xiujuan
Qiao, Jiaqi
Machine Learning
Artificial Intelligence
Computation and Language
Emotion Recognition in Conversation (ERC) is a practical and challenging task. This paper proposes a novel multimodal approach, the Long-Short Distance Graph Neural Network (LSDGNN). Based on the Directed Acyclic Graph (DAG), it constructs a long-distance graph neural network and a short-distance graph neural network to obtain multimodal features of distant and nearby utterances, respectively. To ensure that long- and short-distance features are as distinct as possible in representation while enabling mutual influence between the two modules, we employ a Differential Regularizer and incorporate a BiAffine Module to facilitate feature interaction. In addition, we propose an Improved Curriculum Learning (ICL) to address the challenge of data imbalance. By computing the similarity between different emotions to emphasize the shifts in similar emotions, we design a "weighted emotional shift" metric and develop a difficulty measurer, enabling a training process that prioritizes learning easy samples before harder ones. Experimental results on the IEMOCAP and MELD datasets demonstrate that our model outperforms existing benchmarks.
title Long-Short Distance Graph Neural Networks and Improved Curriculum Learning for Emotion Recognition in Conversation
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2507.15205