World Action Modeling with Progressive Visual Planning

  • 类型：arxiv
  • 标识：2610.02508
  • 链接：https://arxiv.org/abs/2610.02508
  • 主分类：multimodal
  • 形态：method
  • TLDR：World action models (WAMs) have emerged as a promising paradigm for robotic control by jointly predicting future visual dynamics and actions from an initial observation and instruction. However, existing WAMs struggle with long-horizon prediction, as generating dense video rollouts is highly inefficient. Some recent WAMs address this by predicting a single future frame without generating the full video, but this approach neglects how to progress toward the goal. We present ProWAM, a progressive world action model that jointly predicts actions and an ordered sequence of sparse visual sub-goals,
  • 待LLM分类：否
  • 标题中文：World Action Modeling with Progressive Visual Planning：基于渐进视觉规划的世界动作建模
  • TLDR中文：世界动作模型（WAM）已成为一种有前景的机器人控制范式，从初始观测与指令联合预测未来视觉动态与动作。然而，现有 WAM 难以处理长程预测，因为密集生成视频展开的效率极低。一些近期的 WAM 通过仅预测单帧未来画面而非生成完整视频来解决此问题，但忽略了如何向目标推进。我们提出 ProWAM，一种渐进式世界动作模型，联合预测动作与按序排列的稀疏视觉子目标
  • 来源文件：
  • /inbox/tom/_candidates/2026-10-05-agent-rag-longcontext-candidates.json