ImageWAM：世界行动模型真的需要视频生成吗？还是只是需要图像编辑？
ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?
摘要
世界动作模型通常依赖视频生成技术来实现视觉世界建模与机器人控制之间的连接。然而，基于视频的世界动作模型存在三个共同的限制：密集的多帧未来数据使得推理过程成本较高；完整的视频预测会消耗大量资源来处理与动作无关的时间和外观细节；而长距离未来的预测则可能导致错误，从而误导动作预测结果。这些问题引发了这样一个问题：世界动作模型真的需要视频生成技术吗？我们提出了ImageWAM这一简单的世界动作模型框架，该框架利用预训练的图像编辑模型来辅助机器人动作预测。与视频生成技术不同，图像编辑技术能够提供更合适的先验信息：它只需对目标帧进行变换建模，重点关注与动作相关的当前状态与目标状态的视觉差异，并通过编辑训练将任务指令转化为局部视觉变化。在实际应用中，ImageWAM在推理过程中不会解码目标帧，而是将图像编辑过程中的缓存信息作为世界动作模型的上下文信息来使用。ImageWAM在各种模拟器和现实环境实验中都优于传统的VLA模型以及其他具有竞争力的世界动作模型，且所需的计算量仅为基于视频的世界动作模型的1/6，延迟也仅为1/4。注意力分析还表明，图像编辑缓存所关注的区域与任务相关，因此图像编辑技术可以作为基于视频的世界动作建模的有效替代方案。
English Abstract
World Action Models (WAMs) commonly rely on video generation to bridge visual world modeling and robot control. However, video-based WAMs face three coupled limitations: dense multi-frame future tokens make inference costly, full video prediction spends capacity on action-irrelevant temporal and appearance details, and long-horizon future imagination may introduce errors that mislead action prediction. These issues raise a simple question: Does world action model really need video generation? We propose ImageWAM, a simple WAM framework that repurposes pretrained image editing models for robot action prediction. In contrast to video generation, image editing provides a better-matched prior: it only needs to model a target-frame transformation, focuses on action-relevant current-to-target visual differences, and grounds task instructions to localized visual changes through edit pretraining. In practice, ImageWAM does not decode the target frame at inference time; instead, it conditions a flow-matching action expert on the KV caches produced by image-editing denoising, using them as a compact world-action context. ImageWAM outperforms standard VLA baselines and matching competitive WAMs without additional policy pretraining across different simulator and real-world experiments. It also reduces FLOPs to 1/6 and latency to 1/4 of video-based WAMs. Attention analysis further shows that editing caches focus on task-relevant change regions, supporting image editing as an effective alternative to video-based world-action modeling.