AffectAgent: Collaborative Multi-Agent Reasoning for Retrieval-Augmented Multimodal Emotion Recognition

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Zeheng, Yu, Zitong, Zhu, Yijie, Zhao, Bo, Liang, Haochen, Wang, Taorui, Xia, Wei, Zhang, Jiayu, Liu, Zhishu, Ma, Hui, Ma, Fei, Tian, Qi
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917407248875520
author Wang, Zeheng
Yu, Zitong
Zhu, Yijie
Zhao, Bo
Liang, Haochen
Wang, Taorui
Xia, Wei
Zhang, Jiayu
Liu, Zhishu
Ma, Hui
Ma, Fei
Tian, Qi
author_facet Wang, Zeheng
Yu, Zitong
Zhu, Yijie
Zhao, Bo
Liang, Haochen
Wang, Taorui
Xia, Wei
Zhang, Jiayu
Liu, Zhishu
Ma, Hui
Ma, Fei
Tian, Qi
contents LLM-based multimodal emotion recognition relies on static parametric memory and often hallucinates when interpreting nuanced affective states. In this paper, given that single-round retrieval-augmented generation is highly susceptible to modal ambiguity and therefore struggles to capture complex affective dependencies across modalities, we introduce AffectAgent, an affect-oriented multi-agent retrieval-augmented generation framework that leverages collaborative decision-making among agents for fine-grained affective understanding. Specifically, AffectAgent comprises three jointly optimized specialized agents, namely a query planner, an evidence filter, and an emotion generator, which collaboratively perform analytical reasoning to retrieve cross-modal samples, assess evidence, and generate predictions. These agents are optimized end-to-end using Multi-Agent Proximal Policy Optimization (MAPPO) with a shared affective reward to ensure consistent emotion understanding. Furthermore, we introduce Modality-Balancing Mixture of Experts (MB-MoE) and Retrieval-Augmented Adaptive Fusion (RAAF), where MB-MoE dynamically regulates the contributions of different modalities to mitigate representation mismatch caused by cross-modal heterogeneity, while RAAF enhances semantic completion under missing-modality conditions by incorporating retrieved audiovisual embeddings. Extensive experiments on MER-UniBench demonstrate that AffectAgent achieves superior performance across complex scenarios. Our code will be released at: https://github.com/Wz1h1NG/AffectAgent.
format Preprint
id arxiv_https___arxiv_org_abs_2604_12735
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle AffectAgent: Collaborative Multi-Agent Reasoning for Retrieval-Augmented Multimodal Emotion Recognition
Wang, Zeheng
Yu, Zitong
Zhu, Yijie
Zhao, Bo
Liang, Haochen
Wang, Taorui
Xia, Wei
Zhang, Jiayu
Liu, Zhishu
Ma, Hui
Ma, Fei
Tian, Qi
Computer Vision and Pattern Recognition
LLM-based multimodal emotion recognition relies on static parametric memory and often hallucinates when interpreting nuanced affective states. In this paper, given that single-round retrieval-augmented generation is highly susceptible to modal ambiguity and therefore struggles to capture complex affective dependencies across modalities, we introduce AffectAgent, an affect-oriented multi-agent retrieval-augmented generation framework that leverages collaborative decision-making among agents for fine-grained affective understanding. Specifically, AffectAgent comprises three jointly optimized specialized agents, namely a query planner, an evidence filter, and an emotion generator, which collaboratively perform analytical reasoning to retrieve cross-modal samples, assess evidence, and generate predictions. These agents are optimized end-to-end using Multi-Agent Proximal Policy Optimization (MAPPO) with a shared affective reward to ensure consistent emotion understanding. Furthermore, we introduce Modality-Balancing Mixture of Experts (MB-MoE) and Retrieval-Augmented Adaptive Fusion (RAAF), where MB-MoE dynamically regulates the contributions of different modalities to mitigate representation mismatch caused by cross-modal heterogeneity, while RAAF enhances semantic completion under missing-modality conditions by incorporating retrieved audiovisual embeddings. Extensive experiments on MER-UniBench demonstrate that AffectAgent achieves superior performance across complex scenarios. Our code will be released at: https://github.com/Wz1h1NG/AffectAgent.
title AffectAgent: Collaborative Multi-Agent Reasoning for Retrieval-Augmented Multimodal Emotion Recognition
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.12735