arXiv：cs.AI· Gawon Seo, Dongwon Kim, Suha Kwak·· 3 小时前
ACID：用逆动力学实现动作一致性，提升世界模型规划效率
ACID: Action Consistency via Inverse Dynamics for Planning with World Models
AI 导读
ACID 是一个决策时规划框架，通过逆动力学模型从预测的状态转移中反推动作，并要求其与条件动作一致，以循环动作一致性残差检验中间转移的可实现性。该残差经尺度不变的自适应权重融入规划成本。在四类动作条件世界模型和八项任务（涵盖仿真物体操作与关节控制、视觉导航及真机操作）中，ACID 一致提升规划表现，并以显著更少的规划算力匹配基线精度。
正文
Abstract:Decision-time planning with action-conditioned world models has become a popular paradigm for embodied control. However, the standard planning cost judges a candidate solely by how close its predicted terminal state lies to the goal, leaving the realizability of the intermediate transitions unchecked--a predicted trajectory can look convincing while the environment rollout drifts away from it. In this paper, we propose ACID, a decision-time planning framework that introduces cycle action consistency: the action inferred backward from a predicted transition by an inverse dynamics model should recover the one that was conditioned on. We fold this per-step residual into the planning cost via a scale-invariant adaptive weight. Across four action-conditioned world models and eight tasks encompassing object manipulation and articulated control in simulation, visual navigation, and real-robot manipulation, ACID consistently improves planning and matches the baseline's accuracy with substantially less planning compute.
| Comments: | Project page: this https URL |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.02403 [cs.RO] |
| (or arXiv:2607.02403v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2607.02403 arXiv-issued DOI via DataCite |
Submission history
From: Gawon Seo [view email]
[v1]
Thu, 2 Jul 2026 16:38:10 UTC (25,165 KB)
[v2]
Thu, 8 Oct 2026 10:23:17 UTC (27,425 KB)
来源：arXiv：cs.AI · arxiv.org