CauSim: Scaling Causal Reasoning with Increasingly Complex Causal Simulators

2026-05-09

AI Summary

Large language models face limitations in causal reasoning due to the complexity of causal systems and the scarcity of executable ground-truth data. To address this, this work proposes CauSim, a novel framework that, for the first time, automatically constructs verifiable structural causal model (SCM) simulators from non-executable causal knowledge. The approach reframes causal reasoning as a scalable supervised learning problem by leveraging curriculum-based complexity progression, bidirectional translation between natural language and executable models, and domain-informed data augmentation. These mechanisms enable cross-representational generalization and model self-improvement. Experimental results demonstrate that CauSim substantially enhances performance across diverse causal reasoning tasks, empirically validating a positive relationship among simulator complexity, data scale, and model capability.

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.

Problem

• causal reasoning
• large language models
• structural causal models
• data scarcity
• executable representations

Innovation

• Causal Reasoning
• Structural Causal Models
• Large Language Models
• Data Augmentation
• Executable Simulation