PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making
Abstract
Effective extraction of the world knowledge in LLMs for complex decision-making tasks remains a challenge. We propose a framework PIANIST for decomposing the world model into seven intuitive components conducive to zero-shot LLM generation. Given only the natural language description of the game and how input observations are formatted, our method can generate a working world model for fast and efficient MCTS simulation. We show that our method works well on two different games that challenge the planning and decision making skills of the agent for both language and non-language based action taking, without any training on domain-specific training data or explicitly defined world model.
Cite
Text
Light et al. "PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making." NeurIPS 2024 Workshops: LanGame, 2024.Markdown
[Light et al. "PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making." NeurIPS 2024 Workshops: LanGame, 2024.](https://mlanthology.org/neuripsw/2024/light2024neuripsw-pianist/)BibTeX
@inproceedings{light2024neuripsw-pianist,
title = {{PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making}},
author = {Light, Jonathan and Xing, Sixue and Liu, Yuanzhe and Chen, Weiqin and Cai, Min and Chen, Xiusi and Wang, Guanzhi and Cheng, Wei and Yue, Yisong and Hu, Ziniu},
booktitle = {NeurIPS 2024 Workshops: LanGame},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/light2024neuripsw-pianist/}
}