Neural Algorithmic Reasoners informed Large Language Model for Multi-Agent Path Finding

TL;DR

This paper introduces LLM-NAR, a novel framework combining neural algorithmic reasoners with large language models to enhance multi-agent path finding, demonstrating significant performance improvements in simulations and real-world tests.

Contribution

It is the first to integrate neural algorithmic reasoners with LLMs for MAPF, improving planning and coordination capabilities.

Findings

Outperforms existing LLM-based MAPF methods in simulations.

Effectively adapts to various LLM models.

Shows superior results in real-world experiments.

Abstract

The development and application of large language models (LLM) have demonstrated that foundational models can be utilized to solve a wide array of tasks. However, their performance in multi-agent path finding (MAPF) tasks has been less than satisfactory, with only a few studies exploring this area. MAPF is a complex problem requiring both planning and multi-agent coordination. To improve the performance of LLM in MAPF tasks, we propose a novel framework, LLM-NAR, which leverages neural algorithmic reasoners (NAR) to inform LLM for MAPF. LLM-NAR consists of three key components: an LLM for MAPF, a pre-trained graph neural network-based NAR, and a cross-attention mechanism. This is the first work to propose using a neural algorithmic reasoner to integrate GNNs with the map information for MAPF, thereby guiding LLM to achieve superior performance. LLM-NAR can be easily adapted to various LLM models. Both simulation and real-world experiments demonstrate that our method significantly outperforms existing LLM-based approaches in solving MAPF problems.