Symphony-Coord: Adaptive Routing for Multi-Agent LLM Systems
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911734015459328 |
|---|---|
| author | Guan, Zhaoyang Cao, Huixi Zhong, Ming Wang, Yin Liu, Guanyu Yang, Eric Ai, Lynn Ni, Yongxin Shi, Bill |
| author_facet | Guan, Zhaoyang Cao, Huixi Zhong, Ming Wang, Yin Liu, Guanyu Yang, Eric Ai, Lynn Ni, Yongxin Shi, Bill |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00966 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Symphony-Coord: Adaptive Routing for Multi-Agent LLM Systems Guan, Zhaoyang Cao, Huixi Zhong, Ming Wang, Yin Liu, Guanyu Yang, Eric Ai, Lynn Ni, Yongxin Shi, Bill Multiagent Systems 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. |
| title | Symphony-Coord: Adaptive Routing for Multi-Agent LLM Systems |
| topic | Multiagent Systems |
| url | https://arxiv.org/abs/2602.00966 |