让大模型与统计方法协作，更准地发现因果关系。
CauScientist: Teaching LLMs to Respect Data for Causal Discovery
- 用大模型提假设，统计方法严格验证，双引擎协同
- 在37节点图上，结构错误率降低44.0%，召回率达100%
- 适合需要高可靠性因果推断的研究者或工业应用
因果发现是科学理解与可靠决策的基础。现有方法存在显著局限：纯数据驱动方法受统计不可区分性与建模假设制约，而近期基于大模型的方法或忽视统计证据，或引入未经验证的先验导致结果偏差。为此，我们提出CauScientist，一种协同框架，将大模型作为假设生成的“数据科学家”，与概率统计作为严谨的“验证者”结合。CauScientist采用混合初始化选择优质初始图结构，通过大模型提出的修改迭代优化，并利用误差记忆引导高效搜索空间。实验表明，该方法显著优于纯数据驱动基线，在F1分数上提升最高达53.8%，召回率从35.0%提升至100.0%。值得注意的是，尽管独立大模型性能随图复杂度下降，CauScientist相较Qwen3-32B在37节点图上仍使结构哈明距离（SHD）降低44.0%。项目主页见https://github.com/OpenCausaLab/CauScientist。
原文摘要 · Abstract (English)
Causal discovery is fundamental to scientific understanding and reliable decision-making. Existing approaches face critical limitations: purely data-driven methods suffer from statistical indistinguishability and modeling assumptions, while recent LLM-based methods either ignore statistical evidence or incorporate unverified priors that can mislead result. To this end, we propose CauScientist, a collaborative framework that synergizes LLMs as hypothesis-generating "data scientists" with probabilistic statistics as rigorous "verifiers". CauScientist employs hybrid initialization to select superior starting graphs, iteratively refines structures through LLM-proposed modifications validated by statistical criteria, and maintains error memory to guide efficient search space. Experiments demonstrate that CauScientist substantially outperforms purely data-driven baselines, achieving up to 53.8% F1 score improvement and enhancing recall from 35.0% to 100.0%. Notably, while standalone LLM performance degrades with graph complexity, CauScientist reduces structural hamming distance (SHD) by 44.0% compared to Qwen3-32B on 37-node graphs. Our project page is at https://github.com/OpenCausaLab/CauScientist.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。