发表机构
Alibaba Tongyi Lab（阿里巴巴通义实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
WanToFight是用于多人战斗交互的实时生成游戏引擎，基于Wan-1.3B视频扩散变换器构建三个组件，解决了先前引擎未共同处理的问题，能结合多种玩法，在特定配置下维持30FPS，是首个此类集成系统。
AI 中文摘要
我们展示了WanToFight，这是一个生成式游戏引擎，可根据键盘输入模拟实时两人的《拳皇97》游戏玩法。先前的生成式游戏引擎要么针对单人第一人称设置，要么针对非实时合作场景；多人控制、实时推理、复杂的物理交互和对抗性游戏玩法尚未得到共同解决。WanToFight通过基于Wan-1.3B视频扩散变换器构建的三个组件填补了这一空白：具有块因果注意力和滚动KV缓存的流式自回归生成器；一个视觉基础的玩家关联模块，将每个玩家的键盘信号绑定到一个角色身份；以及一个通过单人到全游戏课程训练的门控、局部因果键盘注入模块。一个经过四步DMD提炼的学生与一个剪枝的VAE解码器在单个NVIDIA RTX 5090上以512x384的分辨率在完整比赛期间维持30FPS。据我们所知，WanToFight是第一个在一个系统中结合多人控制、实时推理、复杂物理交互和对抗性游戏玩法的生成式游戏引擎。
英文摘要
We present WanToFight, a generative game engine that simulates real-time, two-player The King of Fighters '97 (KOF~'97) gameplay from keyboard input. Prior generative game engines target either single-player first-person settings or non-real-time cooperative scenarios; multi-player control, real-time inference, complex physical interaction, and adversarial gameplay have not been jointly addressed. WanToFight closes this gap with three components built on the Wan-1.3B video diffusion transformer: a streaming autoregressive generator with block-causal attention and a rolling KV cache; a visually grounded Player Association module that binds each player's keyboard signal to a character identity; and a gated, locally causal keyboard injection module trained with a single-player-to-full-gameplay curriculum. A four-step DMD-distilled student paired with a pruned VAE decoder sustains 30FPS at 512x384 on a single NVIDIA RTX 5090 over the duration of a complete match. To our knowledge, WanToFight is the first generative game engine to combine multi-player control, real-time inference, complex physical interaction, and adversarial gameplay in one system.
CommentsProject Page: https://humanaigc.github.io/wantofight/
