用多专家树状提问法，让大模型透明推断因果关系并给出可信度评分。
Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimation
- 将因果推断拆解为短序列的路径与独立性问题问答，提升可解释性。
- 在无数据基准上优于直接调用大模型，关键因果关系判断准确率更高。
- 适合需要可解释因果推理的科研与医疗场景，尤其关注可信度评估。
因果发现旨在揭示“什么导致什么”，但传统约束方法（如PC、FCI）存在误差传播问题，而现有基于大模型的因果工具常作为不可解释的黑箱，且无置信度输出。本文提出Tree-Query——一种树状结构的多专家大模型框架，将成对因果发现简化为关于后门路径、独立性、潜在混杂和因果方向的一系列简短查询，生成可解释的判断结果，并附带考虑鲁棒性的置信度分数。理论证明了四种成对关系的渐近可识别性。在基于Mooij等人的数据集与UCI因果图构建的无数据基准上，Tree-Query在结构指标上优于直接调用大模型的基线；通过饮食-体重案例研究，展示了混杂因子筛选与稳定高置信度因果结论的生成能力。该方法为从大模型中获取可信赖的数据无关因果先验提供了系统化路径。代码已公开于https://anonymous.4open.science/r/Repo-9B3E-4F96。
原文摘要 · Abstract (English)
Causal discovery aims to recover ``what causes what'', but classical constraint-based methods (e.g., PC, FCI) suffer from error propagation, and recent LLM-based causal oracles often behave as opaque, confidence-free black boxes. This paper introduces Tree-Query, a tree-structured, multi-expert LLM framework that reduces pairwise causal discovery to a short sequence of queries about backdoor paths, (in)dependence, latent confounding, and causal direction, yielding interpretable judgments with robustness-aware confidence scores. Theoretical guarantees are provided for asymptotic identifiability of four pairwise relations. On data-free benchmarks derived from Mooij et al. and UCI causal graphs, Tree-Query improves structural metrics over direct LLM baselines, and a diet--weight case study illustrates confounder screening and stable, high-confidence causal conclusions. Tree-Query thus offers a principled way to obtain data-free causal priors from LLMs that can complement downstream data-driven causal discovery. Code is available at https://anonymous.4open.science/r/Repo-9B3E-4F96.
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。