Multimodal Adaptive Expert Selection with Text Routing and Ordinal Prototype Optimization for Sentiment Analysis
2026-08-31 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors developed a new method called MAESTRO to better understand emotions by combining spoken words, voice tone, and facial expressions. Instead of treating all data the same way, their approach uses the meaning of words to decide which parts of the audio and video to focus on, kind of like a conductor guiding musicians. They also created a way to recognize that emotions have different strengths, making the model more sensitive to subtle changes. Tests show their method improves accuracy in identifying emotions and offers clear reasoning behind its decisions.
Multimodal Sentiment AnalysisMixture-of-ExpertsContrastive LearningOrdinal RegressionAffective ComputingCross-modal FusionDynamic RoutingPrototype LearningCMU-MOSICMU-MOSEI
Authors
Xiaode Chen, Jiakang Yu, Hongtao Deng, Huina Qu, Xun Zhu, Yinxia Lou
Abstract
Multimodal Sentiment Analysis (MSA) is a fundamental component of affective computing that aims to decipher complex emotional states by integrating verbal content with non-verbal cues including vocal intonation and facial micro-expressions. While recent disentanglement-based approaches have advanced the field, their potential is hindered by two methodological challenges. First, static computation graphs process all samples indiscriminately regardless of semantic complexity, which leads to suboptimal representation for diverse emotional expressions and contextual scenarios. Second, generic contrastive objectives often neglect the intrinsic ordinal hierarchy of sentiment intensities. To systematically address these limitations, we introduce Multimodal Adaptive Expert Selection with Text Routing and Ordinal prototype optimization (MAESTRO), a novel framework designed to dynamically orchestrate and refine multimodal representations. Drawing inspiration from an orchestra conductor, we design a Text-Guided Hybrid Mixture-of-Experts (MoE) mechanism. Unlike static fusion, this module utilizes linguistic context as a routing signal to dynamically activate specific audio-visual experts, thereby resolving cross-modal ambiguity through adaptive feature enhancement. Furthermore, to capture fine-grained sentiment gradations, we propose an Ordinal-aware Prototype Contrastive Learning (O-PCL). By incorporating distance-based penalties into the prototype learning objective, O-PCL enforces a structured latent space that preserves the natural order of emotion. Extensive experiments on the CMU-MOSI and CMU-MOSEI benchmarks demonstrate that MAESTRO achieves state-of-the-art performance, and qualitative analysis further confirms the interpretability of our dynamic routing paradigm.