From Word to World: Can Large Language Models be Implicit Text-based World Models?

TL;DR

This paper investigates whether large language models can serve as effective implicit world models in text-based environments, demonstrating their potential to improve agent learning under certain conditions.

Contribution

It introduces a three-level framework for evaluating LLM-based world models and shows their ability to enhance agent performance in specific scenarios.

Findings

LLMs maintain coherent latent states in text environments

World models scale predictably with data and size

They improve agent learning through various methods

Abstract

Agentic reinforcement learning increasingly relies on experience-driven scaling, yet real-world environments remain non-adaptive, limited in coverage, and difficult to scale. World models offer a potential way to improve learning efficiency through simulated experience, but it remains unclear whether large language models can reliably serve this role and under what conditions they meaningfully benefit agents. We study these questions in text-based environments, which provide a controlled setting to reinterpret language modeling as next-state prediction under interaction. We introduce a three-level framework for evaluating LLM-based world models: (i) fidelity and consistency, (ii) scalability and robustness, and (iii) agent utility. Across five representative environments, we find that sufficiently trained world models maintain coherent latent state, scale predictably with data and model size, and improve agent performance via action verification, synthetic trajectory generation, and warm-starting reinforcement learning. Meanwhile, these gains depend critically on behavioral coverage and environment complexity, delineating clear boundry on when world modeling effectively supports agent learning.