Revisiting Multimodal Emotion Recognition in Conversation from the Perspective of Graph Spectrum

arXiv:2404.17862v1 Announce Type: new
Abstract: Efficiently capturing consistent and complementary semantic features in a multimodal conversation context is crucial for Multimodal Emotion Recognition in Conversation (MERC). Existing methods mainly use graph structures to model dialogue context semantic dependencies and employ Graph Neural Networks (GNN) to capture multimodal semantic features for emotion recognition. However, these methods are limited by some inherent characteristics of GNN, such as over-smoothing and low-pass filtering, resulting in the inability to learn long-distance consistency information and complementary …

abstract arxiv consistent context conversation cs.cl dependencies dialogue emotion features gnn graph graph neural networks multimodal networks neural networks perspective recognition semantic spectrum type