TodyComm: Task-Oriented Dynamic Communication for Multi-Round LLM-based Multi-Agent System

2026-02-03

AI Summary

This work addresses the limitations of multi-agent systems based on large language models in dynamic environments, where fixed communication structures often hinder adaptability to role changes, adversarial interference, or communication constraints. To overcome this, the authors propose TodyComm, an algorithm that, for the first time, enables task-oriented dynamic generation of communication topologies. By leveraging policy gradient optimization to maximize task utility, TodyComm adaptively reshapes collaboration structures in each interaction round, balancing performance and communication efficiency. Experimental results across five benchmark tasks demonstrate that TodyComm significantly improves task success rates under dynamic adversarial conditions and strict communication budgets, while maintaining high token efficiency and strong scalability.

Technology Category

• Multiagent Systems: Agent Communication
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Multi-round LLM-based multi-agent systems rely on effective communication structures to support collaboration across rounds. However, most existing methods employ a fixed communication topology during inference, which falls short in many realistic applications where the agents'roles may change \textit{across rounds} due to dynamic adversary, task progression, or time-varying constraints such as communication bandwidth. In this paper, we propose addressing this issue through TodyComm, a \textbf{t}ask-\textbf{o}riented \textbf{dy}namic \textbf{comm}unication algorithm. It produces behavior-driven collaboration topologies that adapt to the dynamics at each round, optimizing the utility for the task through policy gradient. Experiments on five benchmarks demonstrate that under both dynamic adversary and communications budgets, TodyComm delivers superior task effectiveness while retaining token efficiency and scalability.

Problem

• multi-agent system
• dynamic communication
• task-oriented
• communication topology
• LLM-based

Innovation

• task-oriented
• dynamic communication
• multi-agent system
• adaptive topology
• policy gradient