SALA：语义对齐的潜在动作实现具身基础模型双轴扩展
SALA: Semantically Aligned Latent Actions Enable Dual-Axis Scaling of Embodied Foundation Models
AI 导读
arXiv 论文提出 SALA，通过语义对齐潜在动作，利用非配对的人类自我中心视频与机器人视频训练具身基础模型，无需配对数据即可建立跨本体语义对应。实验显示，增加通用人类视频预训练可持续降低下游机器人动作预测误差，增加任务特定人类视频后训练则显著提升真实机器人任务成功率。
正文 · 原文
Abstract:Embodied foundation models hinge on diverse robot interaction data, which are costly to collect. Egocentric human videos offer a complementary source of experience, and latent action models (LAMs) provide a popular means of leveraging these videos for policy supervision. Compared with world-action models (WAMs), LAMs can offer greater training efficiency and potentially lower inference costs by avoiding explicit pixel prediction during policy learning and execution. However, most existing LAMs learn latent actions solely through video reconstruction, which can cause their representations to cluster by visual appearance rather than by the underlying actions. As a result, they struggle to align human actions with their robot counterparts. To bridge the gap, we introduce SALA, which emphasizes establishing the semantic correspondence for latent actions. Without requiring paired data, we use the language descriptions of unpaired human and robot clips to estimate semantic similarities and supervise cross-embodiment correspondence. Specifically, we translate these similarities into soft neighborhood structures that guide human latent actions toward semantically related robot latents, while robot action reconstruction grounds the latent space in control. The resulting shared latent action space provides a unified supervision signal for training an embodied foundation model on both human and robot videos. Notably, latent action supervision enables scaling along two axes: (1) pre-training on increasing amounts of general human video consistently reduces robot action prediction error on downstream tasks, and (2) post-training on increasing amounts of task-specific human video substantially improves real-world robot task success rates.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.17110 [cs.RO] |
| (or arXiv:2611.17110v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17110 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiachun Jin [view email]
[v1]
Thu, 12 Nov 2026 17:04:50 UTC (12,741 KB)
来源：arXiv cs.AI · arxiv.org