从大型语言模型中提取因果关系
Causality Elicitation from Large Language Models
AI总结：
本文提出了一种从大型语言模型中提取因果关系的方法，通过采样文档、提取事件列表、归组标准事件并构建二元向量，最终估计候选因果图。
摘要：
大型语言模型（LLMs）是在大量数据上进行训练，并在其参数中编码知识。我们提出了一种从LLMs中提取因果关系的流程。具体来说，（i）我们从给定主题上从LLMs中采样许多文档，（ii）我们从每个文档中提取一个事件列表，（iii）我们将出现在多个文档中的事件归组为标准事件，（iv）我们为每个文档构建一个二元指示向量，覆盖标准事件，（v）我们使用因果发现方法来估计候选因果图。我们的方法不保证现实中的因果性。相反，它提供了一个框架，以呈现LLMs可以合理假设的因果假设集，作为可检查的变量和候选图集。
英文摘要：
Large language models (LLMs) are trained on enormous amounts of data and encode knowledge in their parameters. We propose a pipeline to elicit causal relationships from LLMs. Specifically, (i) we sample many documents from LLMs on a given topic, (ii) we extract an event list from from each document, (iii) we group events that appear across documents into canonical events, (iv) we construct a binary indicator vector for each document over canonical events, and (v) we estimate candidate causal graphs using causal discovery methods. Our approach does not guarantee real-world causality. Rather, it provides a framework for presenting the set of causal hypotheses that LLMs can plausibly assume, as an inspectable set of variables and candidate graphs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
