MultiGen: Level-Design for Editable Multiplayer Worlds in Diffusion Game Engines

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Po, Ryan, Zhang, David Junhao, Hertz, Amir, Wetzstein, Gordon, Wadhwa, Neal, Ruiz, Nataniel
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910087053836288
author Po, Ryan
Zhang, David Junhao
Hertz, Amir
Wetzstein, Gordon
Wadhwa, Neal
Ruiz, Nataniel
author_facet Po, Ryan
Zhang, David Junhao
Hertz, Amir
Wetzstein, Gordon
Wadhwa, Neal
Ruiz, Nataniel
contents Video world models have shown immense promise for interactive simulation and entertainment, but current systems still struggle with two important aspects of interactivity: user control over the environment for reproducible, editable experiences, and shared inference where players hold influence over a common world. To address these limitations, we introduce an explicit external memory into the system, a persistent state operating independent of the model's context window, that is continually updated by user actions and queried throughout the generation roll-out. Unlike conventional diffusion game engines that operate as next-frame predictors, our approach decomposes generation into Memory, Observation, and Dynamics modules. This design gives users direct, editable control over environment structure via an editable memory representation, and it naturally extends to real-time multiplayer rollouts with coherent viewpoints and consistent cross-player interactions.
format Preprint
id arxiv_https___arxiv_org_abs_2603_06679
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MultiGen: Level-Design for Editable Multiplayer Worlds in Diffusion Game Engines
Po, Ryan
Zhang, David Junhao
Hertz, Amir
Wetzstein, Gordon
Wadhwa, Neal
Ruiz, Nataniel
Artificial Intelligence
Computer Vision and Pattern Recognition
Graphics
Video world models have shown immense promise for interactive simulation and entertainment, but current systems still struggle with two important aspects of interactivity: user control over the environment for reproducible, editable experiences, and shared inference where players hold influence over a common world. To address these limitations, we introduce an explicit external memory into the system, a persistent state operating independent of the model's context window, that is continually updated by user actions and queried throughout the generation roll-out. Unlike conventional diffusion game engines that operate as next-frame predictors, our approach decomposes generation into Memory, Observation, and Dynamics modules. This design gives users direct, editable control over environment structure via an editable memory representation, and it naturally extends to real-time multiplayer rollouts with coherent viewpoints and consistent cross-player interactions.
title MultiGen: Level-Design for Editable Multiplayer Worlds in Diffusion Game Engines
topic Artificial Intelligence
Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2603.06679