[论文] WorldCast: Distributed Multiplayer World Models

论文概要

研究领域: CV 作者: Ziyang Ye, Junchao Huang, Evelyn Zhang, Zhihao Xie, Ruicheng Zhang, Boyao Han, Litao Ban, Ziye Wang, Xinting Hu, Shaoshuai Shi, Zhuotao Tian, Li Jiang 发布时间: 2026-10-08 arXiv: 2610.12412

中文摘要

多人世界模型必须生成独立控制的视图，同时保持玩家和共享环境的一致表征。现有方法大多通过联合多视图生成来协调多个玩家，其成本随玩家数量增加而增长。我们提出 WorldCast——一个分布式多人世界模型，每个玩家运行一个包含视频生成器和状态模型的本地客户端。训练期间利用记录的玩家位置和地图几何，状态模型从生成的视频和控制输入估计玩家位置。客户端交换玩家状态并将其投影到相机对齐的玩家状态场中，指导其他玩家在何处以及如何被渲染。共享场景状态使客户端能复用彼此生成的观测来保持跨视图的场景外观一致性。在 Counter-Strike 2 上的实验展示了 WorldCast 的一致性、实时性能和分布式可扩展性。相机对齐的玩家状态场将玩家渲染率比联合生成方法提升了一个数量级以上，而共享场景状态改善了整个回合的视觉一致性。每个客户端实时运行，仅交换玩家和场景状态，实现了无集中计算瓶颈的可扩展多人生成。图像质量在数小时的 rollout 中保持稳定。

原文摘要

Multiplayer world models must generate independently controlled views with consistent representations of both players and their shared environment. Most existing approaches coordinate multiple players through joint multi-view generation, whose cost grows with each additional player. We present WorldCast, a distributed multiplayer world model in which each player runs a local client comprising a video generator and a state model. Using recorded player positions and map geometry during training, the state model estimates the player's position from generated video and control inputs. Clients exchange player states and project them into camera-aligned player state fields that guide where and how other players are rendered. Shared scene state enables clients to reuse one another's generated obs...

#论文 #arXiv #CV #小凯