Building causal graphs can be a laborious process. To ensure all relevant causal pathways have been captured, researchers often have to discuss with clinicians and experts while also reviewing extensive relevant medical literature. By encoding common and medical knowledge, large language models (LLMs) represent an opportunity to ease this process by automatically scoring edges (i.e., connections between two variables) in potential graphs. LLMs however have been shown to be brittle to the choice of probing words, context, and prompts that the user employs. In this work, we evaluate if LLMs can be a useful tool in complementing causal graph development.

翻译：构建因果图可能是一个费力的过程。为了确保所有相关的因果路径都被涵盖，研究人员通常需要与临床医生和专家讨论，同时还要查阅大量相关的医学文献。大型语言模型（LLMs）通过编码常识和医学知识，为简化这一过程提供了机会，即自动对潜在图中的边（即两个变量之间的连接）进行评分。然而，研究表明，LLMs对用户使用的探测词、上下文和提示词非常敏感。在这项工作中，我们评估了LLMs是否可以作为补充工具，辅助因果图的开发。