In this project, our goal is to determine how to leverage the world-knowledge of pretrained large language models for efficient and robust learning in multiagent decision making. We examine this in a taxi routing and assignment problem where agents must decide how to best pick up passengers in order to minimize overall waiting time. While this problem is situated on a graphical road network, we show that with the proper prompting zero-shot performance is quite strong on this task. Furthermore, with limited fine-tuning along with the one-at-a-time rollout algorithm for look ahead, LLMs can out-compete existing approaches with 50 times fewer environmental interactions. We also explore the benefits of various linguistic prompting approaches and show that including certain easy-to-compute information in the prompt significantly improves performance. Finally, we highlight the LLM's built-in semantic understanding, showing its ability to adapt to environmental factors through simple prompts.
View on arXiv@article{su2025_2502.18822,
title={ Data-Efficient Multi-Agent Spatial Planning with LLMs },
author={ Huangyuan Su and Aaron Walsman and Daniel Garces and Sham Kakade and Stephanie Gil },
journal={arXiv preprint arXiv:2502.18822},
year={ 2025 }
}