发表机构
The University of Texas at Dallas; Technical University of Darmstadt; Swarthmore College; Indiana University School of Medicine; Hessian Center for Artificial Intelligence; German Research Center for AI (DFKI)（德克萨斯大学达拉斯分校; 达姆施塔特工业大学; 斯沃斯莫尔学院; 印第安纳大学医学院; 黑森人工智能中心; 德国人工智能研究中心（DFKI））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出神经符号框架，将LLM作为自适应提议分布，结合其先验知识与数据经验评分，在真实临床数据集上建模APOs，恢复专家验证边并识别新因果关系，为干预提供新见解。
AI 中文摘要
早产、妊娠糖尿病等不良妊娠结局（APOs）对母婴均可能产生长期影响，但其病因仍不明确。该领域的因果发现因数据匮乏和领域知识不完整而极具挑战性，导致纯数据驱动方法失效，且大语言模型（LLM）的输出存在不一致或矛盾问题。本文提出一种用于生成合理因果假设的神经符号框架，该框架迭代结合LLM的广泛先验知识与数据上的经验评分。本方法将LLM视为自适应提议分布，生成的假设会与经验数据进行评分；评分较高的图随后用于更新LLM的上下文，引导后续生成向假设空间中更有前景的区域发展。我们在用于建模APOs及其风险因素的真实临床数据集上评估了该方法，并将结果与专家构建的因果图进行比较。本方法恢复了所有专家验证的边，还识别出专家此前未列出的其他合理因果关系，或可为针对性干预提供新见解。
英文摘要
Adverse Pregnancy Outcomes (APOs) such as preterm birth and gestational diabetes can have long-term consequences for both the mother and child, yet an understanding of their causes remains elusive. Causal discovery in this domain is especially challenging due to a paucity of data and incomplete domain knowledge. As a result, pure data-driven methods fail, and Large Language Model (LLM) outputs remain inconsistent or contradictory. We introduce a neurosymbolic framework for generating plausible causal hypotheses that iteratively combines the broad prior knowledge of LLMs with empirical scoring on data. Our method treats the LLM as an adaptive proposal distribution, generating hypotheses that are scored against empirical data; the resulting high-scoring graphs are then used to update the LLM's context, steering subsequent generations toward more promising regions of the hypothesis space. We evaluate our approach on a real-world clinical dataset for modeling APOs and their risk factors, comparing our results against an expert-constructed causal graph. Our method recovers all expert-validated edges and identifies additional plausible causal relations not previously listed by experts, potentially providing new insights for targeted interventions.
