Large Language Models are Effective Priors for Causal Graph Discovery

TL;DR

This paper explores how Large Language Models can serve as effective priors to enhance causal graph discovery, especially in assessing edge directionality, by proposing metrics, prompting strategies, and integration methods.

Contribution

It introduces a systematic approach to evaluate and incorporate LLM-derived priors into causal discovery algorithms, improving performance on benchmarks.

Findings

LLMs can effectively provide priors for causal graph discovery.

Prompting design impacts the quality of LLM-supplied priors.

LLM priors improve causal discovery, especially for edge directionality.

Abstract

Causal structure discovery from observations can be improved by integrating background knowledge provided by an expert to reduce the hypothesis space. Recently, Large Language Models (LLMs) have begun to be considered as sources of prior information given the low cost of querying them relative to a human expert. In this work, firstly, we propose a set of metrics for assessing LLM judgments for causal graph discovery independently of the downstream algorithm. Secondly, we systematically study a set of prompting designs that allows the model to specify priors about the structure of the causal graph. Finally, we present a general methodology for the integration of LLM priors in graph discovery algorithms, finding that they help improve performance on common-sense benchmarks and especially when used for assessing edge directionality. Our work highlights the potential as well as the shortcomings of the use of LLMs in this problem space.