外观
论文arXiv 2610.1763815 天前
通过寄存器行动
原文链接：Acting Through Registers（arXiv 2610.17638 · 发布日期：2026-10-13）
Acting Through Registers
针对WAM动作相关表征组织不清问题，提出RegisterWAM与FM监督，并在LIBERO-Plus、RoboTwin 2.0和真实操作中验证。
核心要点
- WAM用生成式视觉表征条件化机器人动作，但动作相关信息如何组织仍不清楚。
- 研究发现持久高范数空间离群点，其动作相关性在归一化和K/V投影后仍存在。
- RegisterWAM加入可学习非空间寄存器直接条件化动作，并结合观测FM损失抑制空间离群点。
- 模型在LIBERO-Plus达81.64%成功率、RoboTwin 2.0平均93.0%，并提升真实世界操作表现。
验证范围与局限：验证范围包括LIBERO-Plus、RoboTwin 2.0和真实世界操作任务。
英文摘要 (Abstract)
World Action Models (WAMs) use generative visual representations to condition robot actions, yet how these representations organize action-relevant information remains poorly understood. We identify persistent high-norm spatial outliers in WAM representations and find that their action relevance persists even after normalization and K/V projection attenuate their extreme magnitude. Motivated by this observation, we introduce RegisterWAM, which adds learnable non-spatial registers to the world model and uses them directly for action conditioning. Registers alone do not fully eliminate spatial outliers; pairing them with observation flow-matching (FM) loss supervision suppresses spatial outliers while concentrating high-norm responses in registers. Beyond this reorganization, observation FM loss supervision also improves action learning without registers, motivating an additional noised-observation FM objective that learns current-scene recovery from corrupted inputs. The resulting model achieves 81.64% success on LIBERO-Plus without additional embodied pretraining data or foundation models, reaches 93.0% average success on RoboTwin 2.0, and improves real-world manipulation performance.
作者：Yiwei Zhang, Yantai Yang, Jin Gao, Jin Lou, Andong Chen, Yufei Liu, Boyang Xing, Yuchen Zhu, Weiming Hu, Zhipeng Zhang
来源
本页事实对应的公开页面，以原文为准。
- arXiv 2610.17638：Acting Through Registers访问 2026-10-21