HuggingFace Daily Papers·· 1 天前AI 评分41
EmbodiedSmith：用递归自我改进飞轮在仿真中扩展具身数据
EmbodiedSmith: Scaling Embodied Data through Recursive Self-Improvement Flywheel in Simulation
AI 导读
EmbodiedSmith 通过递归自我改进（RSI）在仿真中规模化生成具身数据，将资产、场景与任务生成统一到一条流水线，并让场景生成预判下游任务需求、任务生成反过来引导场景定向编辑。该框架支持移动操作臂、人形机器人和灵巧手，以及可变形物体与流体的交互。下游策略实验表明，数据多样性提升可改善泛化能力。
正文
Authors:Yikai Qin, Yifei Deng, Mingjian Liang, Wenxuan Song, Zepeng Lin, Zhiyi Jiang, Jiajun Fu, Qiao Sun, Huashuo Lei, Xicheng Gong, Jiayi Chen, Han Zhao, Shuanghao Bai, Pengxiang Ding, Pengwei Wang, Haoang Li
Abstract:Scaling robotic foundation models requires diverse training data and reliable evaluation environments. Simulation offers a scalable solution, yet existing generation pipelines remain constrained by predefined assets and skills, a disconnect between scene generation and task generation, and limited support for complex embodiments and physics. We introduce EmbodiedSmith, a framework for scalable embodied data generation through recursive self-improvement (RSI). EmbodiedSmith unifies asset, scene, and task generation in a pipeline that supports autonomous creation and language-driven customization. Its core is an agentic refinement loop: scene generation anticipates downstream task requirements, while task generation guides targeted scene edits, allowing scenes and tasks to iteratively improve one another. This joint refinement improves task generation success, including for long-horizon tasks. The framework further supports mobile manipulators, humanoids, and dexterous hands, as well as interactions involving deformable objects and fluids, broadening the range of behaviors and physical phenomena represented in generated data. Together, these capabilities provide a flexible simulation engine for both robot pretraining and evaluation. Extensive experiments validate the quality, diversity, and generation efficiency of the resulting data, while downstream policy experiments demonstrate that increased data diversity improves generalization.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO) |
| Cite as: | arXiv:2610.07969 [cs.CV] |
| (or arXiv:2610.07969v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07969 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qin Yikai [view email]
[v1]
Tue, 6 Oct 2026 08:34:28 UTC (61,964 KB)
来源：HuggingFace Daily Papers · arxiv.org