CauSim:用日益复杂的因果模拟器扩展大语言模型因果推理
Scaling Causal Reasoning with Increasingly Complex Causal Simulators
AI 导读
研究者提出 CauSim 框架,把大语言模型的因果推理从稀缺标注问题转化为可扩展的监督学习问题。该框架由 LLM 增量构建可执行的结构因果模型(SCM)作为日益复杂的因果模拟器,在扩展到全局复杂系统的同时保持因果查询的可验证答案,并支持将不可执行的因果知识形式化为代码、把 SCM 转译为自然语言。研究表明该方法带来跨表示的泛化、课程式扩展和数据量增加的一致收益,以及通过自生成模拟器实现自我改进。
正文 · 原文
Abstract:Despite surpassing human performance across mathematics, coding, and other knowledge-intensive tasks, large language models (LLMs) continue to struggle with causal reasoning. A core obstacle is the target data itself: causal systems are complex and often expressed in non-executable forms, while ground-truth answers to causal queries are inherently scarce. We introduce CauSim, a framework that turns causal reasoning from a scarce-label problem into a scalable supervised one. CauSim constructs increasingly complex causal simulators: executable structural causal models (SCMs), incrementally built by LLMs, that scale to globally complex systems while maintaining verifiable answers to causal queries. CauSim operates across representations by formalizing non-executable causal knowledge into code, enabling data augmentation, and translating executable SCMs into natural language, enabling supervision in previously difficult-to-supervise representations. We structure our research into two parts: (1) how to construct increasingly complex causal simulators, and (2) a systematic study of what CauSim enables, demonstrating generalization across representations, consistent gains from curriculum scaling and data volume, LLM self-improvement through self-generated simulators, and data augmentation via formalization of existing domain knowledge.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.09079 [cs.AI] |
| (or arXiv:2605.09079v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2605.09079 arXiv-issued DOI via DataCite |
Submission history
From: Nicolas Astorga [view email]
[v1]
Sat, 9 May 2026 17:39:26 UTC (661 KB)
[v2]
Tue, 17 Nov 2026 00:54:34 UTC (2,200 KB)
来源:arXiv cs.AI · arxiv.org