arXiv cs.AI· Juntao Cheng, Jinzhou Tang, Fan Feng, Henrik I. Christensen, Yujia Zheng·· 1 天前AI 评分34
CausalWAM：让世界动作模型学会动作的因果来源
CausalWAM: Learning the Causes of Actions in World Action Models
AI 导读
研究者提出 CWAM（Causal WAM），通过在预训练世界动作模型（WAM）的动作头上增加稀疏路由路径，将稀疏性约束引入模型，使生成动作所用的隐变量仅为动作相关隐变量的可逆函数，且世界建模目标保持不变。理论证明该结构在仅差一个隐变量置换的意义下可被识别，无需充分时间依赖或对未观测状态演化的假设。合成数据与真实机器人实验验证了理论，并显示闭环任务成功率效率提升。
正文 · 原文
Abstract:World action models (WAMs) learn latent variables from observations and use these to both predict the evolution of the scene and to generate actions. The prediction has to be of the whole scene, whereas only a portion of the scene causes actions. WAMs fitted to demonstrations do not, in general, guarantee that they will generate actions solely from the latent causes of the demonstrated actions (the action-relevant latent variables), as the model could just as well learn to imitate in this way whilst entangling the action-relevant latent variables with the nuisance variables. We therefore separate out what causes action from what prediction depends on. We prove that under a sparsity constraint the latent variables used to generate actions are an invertible function of the action-relevant latent variables alone and the true structure is identified up to a permutation of the latent variables. This requires neither sufficient temporal dependence nor any assumptions on how the unobserved regimes evolve and holds even when the action-relevant latent variables are statistically dependent on the rest. We therefore propose CWAM (Causal WAM), which incorporates the constraint into a pretrained WAM by adding a sparse routing path to its action head, with the world modeling objective remaining unchanged. Experiments on both synthetic data and real-world robots validate the theory as well as showcase the improved efficiency in closed-loop success.
| Subjects: | Artificial Intelligence (cs.AI); Robotics (cs.RO); Machine Learning (stat.ML) |
| Cite as: | arXiv:2610.42096 [cs.AI] |
| (or arXiv:2610.42096v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.42096 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yujia Zheng [view email]
[v1]
Fri, 30 Oct 2026 17:40:00 UTC (2,226 KB)
来源：arXiv cs.AI · arxiv.org