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 adversarial settings and communication budget constraints, TodyComm achieves superior task performance while maintaining token efficiency, scalability, and strong generalizability across varying adversarial conditions.

翻译：基于大语言模型（LLM）的多轮多智能体系统需要依赖有效的通信结构来支持跨轮协作。然而，现有方法大多在推理阶段采用固定的通信拓扑，这在许多实际应用中存在局限性——由于动态对手、任务进展或通信带宽等时变约束，智能体的角色可能跨轮发生改变。本文提出通过TodyComm（一种面向任务的动态通信算法）解决该问题。该算法生成行为驱动的协作拓扑，能够自适应每轮的动态变化，并通过策略梯度优化任务效用。在五个基准测试上的实验表明，在动态对抗环境和通信预算约束下，TodyComm在保持令牌效率、可扩展性和跨不同对抗条件的强泛化能力的同时，实现了优异的任务性能。