DAIN: Dynamic Agent-Based Interaction Network for Efficient and Collaborative Multimodal Reasoning

2026-06-29

AI Summary

Existing static multimodal fusion approaches struggle to achieve efficient and adaptive collaborative reasoning in complex scenarios. This work proposes a dynamic multi-agent fusion framework that models multimodal interaction as a context-aware collaborative process: a meta-controller dynamically schedules specialized interaction agents, while a sparse communication mechanism compresses information exchange to reach consensus. By integrating a dynamically sparsely activated architecture with a multi-objective joint optimization loss function, the method enhances both model performance and interpretability while improving computational efficiency. Experimental results demonstrate that the proposed model achieves new state-of-the-art performance across five benchmarks, including a 2.6% accuracy gain on the ADNI dataset, validating the effectiveness of the dynamic scheduling and sparse communication mechanisms.

Abstract

Current multimodal fusion approaches, particularly those based on static Mixture-of-Experts (MoE) architectures, often struggle to provide the adaptive and efficient collaborative reasoning required by complex real-world applications. We introduce the Dynamic Agent-based Interaction Network (DAIN), which reconceptualizes multimodal fusion as a dynamic, multi-agent collaborative process. DAIN employs a context-aware Meta-Controller that dynamically schedules sparse activation of specialized interaction agents and orchestrates compressed inter-agent communication for consensus-building. The framework is guided by a multi-objective loss function that jointly optimizes task accuracy, agent specialization, and operational efficiency through sparse activation and communication regularization. Comprehensive evaluations across five diverse benchmarks -- ADNI, MIMIC-IV, MM-IMDB, CMU-MOSI, and ENRICO -- establish DAIN as a new state-of-the-art, delivering significant performance improvements including a 2.6\% accuracy gain on ADNI. Ablation studies verify the critical roles of both dynamic scheduling and agent communication. Furthermore, DAIN offers enhanced interpretability by exposing context-dependent agent roles and collaboration patterns while maintaining computational efficiency through sample-wise sparse agent activation. Our work demonstrates the promise of dynamic, agent-based paradigms for multimodal reasoning.

Problem

• multimodal fusion
• static Mixture-of-Experts
• adaptive reasoning
• efficient collaboration
• complex real-world applications

Innovation

• Dynamic Agent-based Interaction
• Sparse Activation
• Multimodal Fusion
• Context-aware Meta-Controller
• Collaborative Reasoning