arXiv cs.AI· Jialei Chen, Kai Wang, Kang Chen, Shuaihang Chen, Feng Gao, Wenhao Tang, Zhiyuan Li, Weilin Liu, Zhuyu Yao, Boxun Li, Yuanbo Xu, Chao Yu·· 4 小时前AI 评分46
LaWAM：基于潜在世界动作模型的高效动力学感知机器人策略
LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies
AI 导读
研究者提出 LaWAM，一种潜在世界动作模型，通过紧凑的潜在视觉子目标而非重建未来视频，将预测动力学引入机器人策略。其核心是潜在动作条件化的潜在世界模型 LaWM。LaWAM 在 LIBERO 上达 98.6% 成功率、RoboTwin 上达 91.22%，每次动作块预测仅需 187 ms，推理延迟最高比像素级 WAM 低 24 倍，已被 CoRL 2026 接收。
正文 · 原文
Authors:Jialei Chen, Kai Wang, Kang Chen, Shuaihang Chen, Feng Gao, Wenhao Tang, Zhiyuan Li, Weilin Liu, Zhuyu Yao, Boxun Li, Yuanbo Xu, Chao Yu
Abstract:Vision-Language-Action models (VLAs) leverage large-scale vision-language pretraining for semantic robot control, but often lack explicit foresight into how robot actions change the scene. World-Action Models (WAMs) address this limitation by conditioning policies on predicted futures, yet existing approaches typically rely on computationally expensive video generation with substantial pixel-level redundancy. We present LaWAM, a Latent World Action Model that exposes predictive dynamics to robot policies through compact latent visual subgoals instead of reconstructed future video. At the core of LaWAM is a latent-action-conditioned Latent World Model (LaWM). We obtain LaWM by training a latent action model in the latent space of a pretrained vision foundation model and repurposing its forward decoder to predict future observation features for scene evolution. LaWAM then conditions action generation on these predicted latent visual subgoals to enable dynamics-aware robot control. LaWAM achieves state-of-the-art or competitive success rates (SRs) across LIBERO (98.6% SR), RoboTwin (91.22% SR), and real-world manipulation tasks while retaining low-latency inference. LaWAM runs in 187 ms per action-chunk prediction and achieves up to 24x lower wall-clock latency than pixel-space WAMs.
| Comments: | Accepted at CoRL 2026. Project website: this https URL |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.15768 [cs.RO] |
| (or arXiv:2606.15768v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2606.15768 arXiv-issued DOI via DataCite |
Submission history
From: Jialei Chen [view email]
[v1]
Sun, 14 Jun 2026 12:06:58 UTC (8,833 KB)
[v2]
Sun, 25 Oct 2026 02:54:12 UTC (21,242 KB)
来源：arXiv cs.AI · arxiv.org