TraceGen：在3D轨迹空间中进行世界建模，实现跨具身视频学习
TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment Videos
把世界模型的预测空间从像素空间换成紧凑的3D轨迹（trace）空间，用跨具身视频预训练出可迁移的3D运动先验，从而在只有5条目标机器人视频时也能少样本适配；对做视频世界模型、机器人策略学习和跨域表征的人来说，这是一个可复用的「用几何轨迹替代像素生成」的建模思路。
原文依据 a world model that predicts future motion in trace-space rather than pixel space, abstracting away appearance while retaining the geometric structure needed for manipulation
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论仅凭少量演示就在新平台、新场景上学习新的机器人任务仍然具有挑战性。尽管其他具身形态（人类与不同机器人）的视频十分丰富，但具身形态、相机与环境上的差异阻碍了它们的直接使用。我们通过引入一种统一的符号化表示——场景级轨迹的紧凑3D「轨迹空间」——来解决小数据问题，该表示使得能够从跨具身、跨环境、跨任务的视频中学习。我们提出 TraceGen，一个在轨迹空间而非像素空间中预测未来运动的世界模型，它抽象掉外观信息，同时保留操作所需的几何结构。为大规模训练 TraceGen，我们构建了 TraceForge，一个将异构的人类与机器人视频转换为一致的3D轨迹的数据流水线，产出了包含123K段视频、1.8M个「观测—轨迹—语言」三元组的数据语料。在该语料上预训练可得到一个可迁移的3D运动先验，其适配十分高效：仅用五个目标机器人视频，TraceGen 在四项任务上即达到80%的成功率，同时推理速度比最优的基于视频的世界模型快50–600倍。在更具挑战性的情形下——即仅有五段用手机拍摄、未经标定的人类演示视频——它仍能在真实机器人上达到67.5%的成功率，凸显了 TraceGen 无需繁重的像素空间生成即可跨具身形态适配的能力。
英文原文摘要 1467 字符
Learning new robot tasks on new platforms and in new scenes from only a handful of demonstrations remains challenging. While videos of other embodiments---humans and different robots---are abundant, differences in embodiment, camera, and environment hinder their direct use. We address the small-data problem by introducing a unifying, symbolic representation---a compact 3D "trace-space" of scene-level trajectories---that enables learning from cross-embodiment, cross-environment, and cross-task videos. We present TraceGen, a world model that predicts future motion in trace-space rather than pixel space, abstracting away appearance while retaining the geometric structure needed for manipulation. To train TraceGen at scale, we develop TraceForge, a data pipeline that transforms heterogeneous human and robot videos into consistent 3D traces, yielding a corpus of 123K videos and 1.8M observation--trace--language triplets. Pretraining on this corpus produces a transferable 3D motion prior that adapts efficiently: with just five target robot videos, TraceGen attains 80% success across four tasks while offering 50-600x faster inference than state-of-the-art video-based world models. In the more challenging case where only five uncalibrated human demonstration videos captured on a handheld phone are available, it still reaches 67.5% success on a real robot, highlighting TraceGen's ability to adapt across embodiments without heavy pixel-space generation.