Causal Inference with Large Language Model: A Survey
Jing Ma Department of Computer and Data Sciences, Case Western Reserve University
Abstract
Causal inference has been a pivotal challenge across diverse domains such as medicine and economics, demanding a complicated integration of human knowledge, mathematical reasoning, and data mining capabilities. Recent advancements in natural language processing (NLP), particularly with the advent of large language models (LLMs), have introduced promising opportunities for traditional causal inference tasks. This paper reviews recent progress in applying LLMs to causal inference, encompassing various tasks spanning different levels of causation. We summarize the main causal problems and approaches, and present a comparison of their evaluation results in different causal scenarios. Furthermore, we discuss key findings and outline directions for future research, underscoring the potential implications of integrating LLMs in advancing causal inference methodologies.
中文速览
因果推断（causal inference）长期依赖人类专家知识与统计方法的结合，但在处理自然语言、融合常识和跨领域泛化时仍力不从心；这篇综述系统梳理了将大型语言模型（LLMs）用于因果推断的最新进展，覆盖因果发现、因果效应估计、反事实推理等多个层次的任务。作者从提示工程、微调、结合传统因果方法、知识增强等技术路径出发，对现有方法进行了分类整理，并横向比较了不同模型在各类因果场景下的评估表现。研究发现，LLMs 在因果常识推理上已展现出相当潜力，但在精确数值估计和保持因果逻辑一致性方面仍有明显短板。该综述为研究者提供了一份清晰的方法全景图，对推动 LLMs 与因果推断深度融合、并将其应用于医疗、金融等高风险领域具有重要参考价值。
原文 arXiv:2409.09822;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2409.09822v3