From Masks to Worlds: A Hitchhiker's Guide to World Models
This offers a practical roadmap for researchers and practitioners interested in developing world models, though it is incremental as it synthesizes existing concepts rather than introducing new methods.
The paper provides a focused guide for building world models by tracing a specific development path from masked models to memory-augmented systems, emphasizing generative, interactive, and memory components as the most promising approach.
This is not a typical survey of world models; it is a guide for those who want to build worlds. We do not aim to catalog every paper that has ever mentioned a ``world model". Instead, we follow one clear road: from early masked models that unified representation learning across modalities, to unified architectures that share a single paradigm, then to interactive generative models that close the action-perception loop, and finally to memory-augmented systems that sustain consistent worlds over time. We bypass loosely related branches to focus on the core: the generative heart, the interactive loop, and the memory system. We show that this is the most promising path towards true world models.