Symphony-Coord: Emergent Coordination in Decentralized Agent Systems

TL;DR

Symphony-Coord introduces a decentralized, adaptive coordination framework for multi-agent systems that organically emerges roles through online learning, improving efficiency and robustness without relying on static roles or centralized control.

Contribution

It transforms agent coordination into an online multi-armed bandit problem with a novel two-stage beacon protocol, enabling scalable, adaptive, and fault-tolerant multi-agent task routing.

Findings

Enhances task routing efficiency in multi-agent systems.

Demonstrates robustness to agent failures and distribution shifts.

Achieves near-optimal allocation with sublinear regret bounds.

Abstract

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 lead to inefficient routing, poor adaptability, and fragile fault recovery capabilities. We introduce Symphony-Coord, a decentralized multi-agent framework that transforms agent selection into an online multi-armed bandit problem, enabling roles to emerge organically through interaction. The framework employs a two-stage dynamic beacon protocol: (i) a lightweight candidate screening mechanism to limit communication and computational overhead; (ii) an adaptive LinUCB selector that routes subtasks based on context features derived from task requirements and agent states, continuously optimized through delayed end-to-end feedback. Under standard linear realizability assumptions, we provide sublinear regret bounds, indicating the system converges toward near-optimal allocation schemes. Validation through simulation experiments and real-world large language model benchmarks demonstrates that Symphony-Coord not only enhances task routing efficiency but also exhibits robust self-healing capabilities in scenarios involving distribution shifts and agent failures, achieving a scalable coordination mechanism without predefined roles.