PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making
This addresses the problem of leveraging LLMs for multi-agent decision-making in partially observable environments, offering a novel approach but with incremental improvements in model generation.
The authors tackled the challenge of extracting world knowledge from LLMs for complex decision-making by proposing PIANIST, a framework that decomposes world models into seven components for zero-shot generation, enabling working world models for MCTS simulation in games without domain-specific training.
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.