Symphony-Coord: Adaptive Routing for Multi-Agent LLM Systems
This work addresses the problem of inefficient and brittle coordination in multi-agent LLM systems, which is important for scaling such systems to dynamic, real-world tasks.
Symphony-Coord introduces a decentralized coordination framework for multi-agent LLM systems that treats agent selection as an online multi-armed bandit problem, enabling adaptive routing without static role assignments. Experiments show improved routing efficiency and fault recovery under distribution shifts and agent failures.
Multi-agent large language model systems can tackle complex multi-step tasks by decomposing work and coordinating specialized behaviors. However, current coordination mechanisms typically rely on statically assigned roles and centralized controllers. As agent pools and task distributions evolve, these design choices can lead to inefficient routing, poor adaptability, and fragile fault recovery. We introduce Symphony-Coord, a task-local coordination framework with decentralized execution that transforms agent selection into an online multi-armed bandit problem. Instead of relying on a fixed task-to-role map, Symphony-Coord allows routing specializations to emerge from interaction and feedback. The framework employs a two-stage dynamic beacon protocol:(i) a lightweight candidate screening mechanism to limit communication and computation overhead; and (ii) an adaptive LinUCB selector that routes subtasks using context features derived from task requirements and agent states, updated through delayed post-execution feedback. Under candidate-conditional linear bandit assumptions, we prove sublinear regret bounds for the immediate-feedback selector and explicitly separate the deferred-update effects introduced by post-vote rewards. Validation through simulation experiments and real-world large language model benchmarks shows that Symphony-Coord improves task routing efficiency and recovery behavior under distribution shifts and agent failures.