Optimal-Agent-Selection: State-Aware Routing Framework for Efficient Multi-Agent Collaboration

TL;DR

This paper introduces STRMAC, a state-aware routing framework for multi-agent systems powered by LLMs, which adaptively selects agents for collaboration, significantly improving efficiency and performance on reasoning tasks.

Contribution

The paper presents a novel state-aware routing method and a self-evolving data generation approach to enhance multi-agent collaboration efficiency and effectiveness.

Findings

Achieves up to 23.8% performance improvement over baselines.

Reduces data collection overhead by up to 90.1%.

Demonstrates state-of-the-art results on collaborative reasoning benchmarks.

Abstract

The emergence of multi-agent systems powered by large language models (LLMs) has unlocked new frontiers in complex task-solving, enabling diverse agents to integrate unique expertise, collaborate flexibly, and address challenges unattainable for individual models. However, the full potential of such systems is hindered by rigid agent scheduling and inefficient coordination strategies that fail to adapt to evolving task requirements. In this paper, we propose STRMAC, a state-aware routing framework designed for efficient collaboration in multi-agent systems. Our method separately encodes interaction history and agent knowledge to power the router, which adaptively selects the most suitable single agent at each step for efficient and effective collaboration. Furthermore, we introduce a self-evolving data generation approach that accelerates the collection of high-quality execution paths for efficient system training. Experiments on challenging collaborative reasoning benchmarks demonstrate that our method achieves state-of-the-art performance, achieving up to 23.8% improvement over baselines and reducing data collection overhead by up to 90.1% compared to exhaustive search.