多人世界模型：把游戏引擎的招数教给AI
现在的视频世界模型一进多人场景就乱套：它把“世界本身”和“某个玩家看到的画面”搅在一起，导致每个视角都要重算一遍、互相还对不上。这篇论文借了多人游戏服务器的思路，把两者拆开：一个“逻辑引擎”只负责推进全局状态（比如蛇的位置、长度），另一个“渲染引擎”按需给每个玩家生成画面。结果在多人贪吃蛇上，状态更准、视角更一致，还能同时模拟1024个玩家跑1万步。它不是你明天能用上的东西，但这是让AI学会“多人世界”的关键一步——就像游戏从单机进化到网游那样。
📄 原文摘要(英文)
Current video world models struggle in multiplayer environments because they entangle world state with view-dependent visual latents, leading to redundant compute, view inconsistencies, and poor scalability. We propose MAS (Multiplayer world models with Authoritative Shared State) to resolve this limitation. Inspired by multiplayer game architectures, MAS disentangles world dynamics and view rendering. A learned Logic Engine advances a global, authoritative typed state from joint actions without any hand-written transition function, acting as the sole recurrent memory and synchronization reference. From this shared state, a learned Rendering Engine generates independent and consistent views for any requested camera on demand. This explicit disentangling allows MAS to achieve superior state accuracy and lower cross-view inconsistency compared to state-of-the-art multi-view baselines on a matched multiplayer Snake benchmark. It advances predicted worlds with 1,024 concurrent players for 10,000 recurrent steps. Our results show that explicit, authoritative state modeling provides a practical foundation for scalable and consistent multi-agent world simulation.