Graph World Models: Concepts, Taxonomy, and Future Directions

TL;DR

This paper introduces graph world models (GWMs), formalizes their taxonomy based on relational inductive biases, and discusses their design principles, challenges, and future research directions.

Contribution

It systematically unifies and surveys emerging graph-based world models under a new formal framework and taxonomy.

Findings

Categorizes GWMs into spatial, physical, and logical RIBs.

Summarizes key design principles and representative models for each category.

Discusses open challenges like dynamic graph adaptation and benchmarks.

Abstract

As one of the mainstream models of artificial intelligence, world models allow agents to learn the representation of the environment for efficient prediction and planning. However, classical world models based on flat tensors face several key problems, including noise sensitivity, error accumulation and weak reasoning. To address these limitations, many recent studies use graph structure to decompose the environment into entity nodes and interactive edges, and model virtual environments in a structured space. This paper systematically formalizes and unifies these emerging graph-based works under the concept of graph world models (GWMs). To the best of our knowledge, GWMs have not yet been explicitly defined and surveyed as a unified research paradigm. Furthermore, we propose a taxonomy based on relational inductive biases (RIB), categorizing GWMs by the specific structural priors they inject: (1) spatial RIB for topological abstraction; (2) physical RIB for dynamic simulation; and (3) logical RIB for causal and semantic reasoning. For each model category, we outline the key design principles, summarize representative models, and conduct comparative analyses. We further discuss open challenges and future directions, including dynamic graph adaptation, probabilistic relational dynamics, multi-granularity inductive biases, and the need for dedicated benchmarks and evaluation metrics for GWMs.