SYMPHONY: Synergistic Multi-agent Planning with Heterogeneous Language Model Assembly

2026-01-30

AI Summary

This work addresses the limited planning performance of conventional single-agent Monte Carlo Tree Search (MCTS), which suffers from insufficient search diversity due to its homogeneous reasoning paradigm. To overcome this limitation, we propose a multi-agent collaborative planning framework that, for the first time, integrates heterogeneous large language models (LLMs) into MCTS to enhance exploration through diversified reasoning strategies. Our approach combines open-source and cloud-based LLMs via a hybrid deployment strategy, achieving significant performance gains over state-of-the-art methods across multiple benchmark tasks. Notably, even when restricted to open-source models executable on consumer-grade hardware, the framework demonstrates consistently strong results, highlighting its practicality and scalability.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Recent advancements have increasingly focused on leveraging large language models (LLMs) to construct autonomous agents for complex problem-solving tasks. However, existing approaches predominantly employ a single-agent framework to generate search branches and estimate rewards during Monte Carlo Tree Search (MCTS) planning. This single-agent paradigm inherently limits exploration capabilities, often resulting in insufficient diversity among generated branches and suboptimal planning performance. To overcome these limitations, we propose Synergistic Multi-agent Planning with Heterogeneous langauge model assembly (SYMPHONY), a novel multi-agent planning framework that integrates a pool of heterogeneous language model-based agents. By leveraging diverse reasoning patterns across agents, SYMPHONY enhances rollout diversity and facilitates more effective exploration. Empirical results across multiple benchmark tasks show that SYMPHONY achieves strong performance even when instantiated with open-source LLMs deployable on consumer-grade hardware. When enhanced with cloud-based LLMs accessible via API, SYMPHONY demonstrates further improvements, outperforming existing state-of-the-art baselines and underscoring the effectiveness of heterogeneous multi-agent coordination in planning tasks.

Problem

• multi-agent planning
• large language models
• Monte Carlo Tree Search
• heterogeneous agents
• exploration diversity

Innovation

• multi-agent planning
• heterogeneous LLMs
• Monte Carlo Tree Search
• reasoning diversity
• synergistic coordination