arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.08260cs.RO
收藏

Seed2Scale: 一种通过小到大模型协同和多模态评估的自我进化数据引擎用于具身AI

Seed2Scale: A Self-Evolving Data Engine with Parallel Worlds Expansion for Scalable Robot Learning

ZTE Corporation(中兴通讯公司)
Cong Tai, Zhaoyu Zheng, Haixu Long, Hansheng Wu, Jialiang Wang, Zhengbin Long, Haodong Xiang, Shuyuan Wang, Rong Shi, Zhuo Cui, Shizhuang Zhang, He Wang, Biao L… 展开作者

Cong Tai, Zhaoyu Zheng, Haixu Long, Hansheng Wu, Jialiang Wang, Zhengbin Long, Haodong Xiang, Shuyuan Wang, Rong Shi, Zhuo Cui, Shizhuang Zhang, He Wang, Biao Liu, Zhenzhe Sun, Tao Shen

AI总结:

Seed2Scale通过小到大模型协同和多模态评估,实现具身AI的自我进化数据引擎,显著提升性能并提供可扩展的开发路径

摘要:

现有数据生成方法存在探索限制、具身差距和低信噪比的问题,导致在自我迭代过程中性能下降。为解决这些挑战,我们提出了Seed2Scale,一种自我进化数据引擎,通过

英文摘要:

Existing data generation methods for robot learning suffer from limited exploration, embodiment gaps, low signal-to-noise ratios, and domain shifts, leading to performance degradation during self-iteration and poor generalization to unseen scenes. To address these challenges, we propose Seed2Scale, a self-evolving data engine with parallel worlds expansion. Starting with as few as four seed demonstrations, Seed2Scale first executes a self-evolution stage driven by a heterogeneous synergy of "small-model collection, large-model evaluation, and target-model learning". Specifically, the lightweight Vision-Language-Action (VLA) model, SuperTiny, serves as a dedicated data collector for robust exploration. Concurrently, a pretrained Vision-Language Model (VLM) functions as a verifier to autonomously score and filter trajectories, supporting stable self-evolution in the evaluated tasks without performance collapse. Furthermore, Seed2Scale introduces a parallel worlds stage, projecting self-evolved trajectories into different environments of the same task to generate more diverse data and enhance adaptability to unseen scenes, including real-world environments. Experimental results demonstrate that Seed2Scale exhibits significant scaling potential: as iterations progress, the success rate of the target model shows a consistent upward trend, significantly outperforming the seed baseline. Notably, Seed2Scale achieves a remarkable 75.38% success rate in zero-shot real-world evaluations, where baseline methods fail completely (0%). Project page: https://terminators2025.github.io/Seed2Scale.github.io

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑