何时因果世界模型才能帮助模块化 LLM 智能体
When Do Causal World Models Help Modular LLM Agents
AI 导读
论文提出 CausalCompose，一个为模块化 LLM 智能体引入因果结构的因果世界模型框架，用于干预感知预测与规划。作者理论上证明，纯观测式世界模型在模块化系统中会产生不可消除的干预时误差，而因果组合在证据充分且局部误差可控时可弥合这一干预差距。跨多种模块化智能体环境的实验显示，因果结构持续提升干预敏感的推理与规划，在因果依赖直接支配动作执行的结构化工具使用系统中收益尤其明显。
正文 · 原文
Abstract:LLM agents increasingly operate in modular systems, where planning requires coordinating multiple functional components whose actions can causally change downstream states and behaviors. This makes intervention-aware prediction essential: an agent must anticipate not only what is likely to happen next, but what will happen as a consequence of its own actions. We find that purely observational world models can incur irreducible intervention-time error in such systems, even when they accurately model observational trajectories. To address this limitation, we propose CausalCompose, a causal world-model framework that equips modular LLM agents with causal structure for intervention-aware prediction and planning. We theoretically show that causal composition can overcome this intervention gap with sufficient evidence and controlled local error. Experiments across diverse modular agent environments validate these results: incorporating causal structure consistently improves intervention-sensitive reasoning and planning, with particularly clear gains in structured tool-use systems where causal dependencies directly govern action execution. Our results show when causal world models help modular LLM agents: when causal dependencies are identifiable and actionable for planning.
| Comments: | Under Review |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.00012 [cs.AI] |
| (or arXiv:2610.00012v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00012 arXiv-issued DOI via DataCite |
Submission history
From: Zekun Cai [view email]
[v1]
Sun, 12 Jul 2026 04:03:37 UTC (2,830 KB)
[v2]
Thu, 29 Oct 2026 13:53:05 UTC (3,732 KB)
来源：arXiv cs.AI · arxiv.org