Prompting Strategies for Enabling Large Language Models to Infer Causation from Correlation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Sgouritsa, Eleni, Aglietti, Virginia, Teh, Yee Whye, Doucet, Arnaud, Gretton, Arthur, Chiappa, Silvia
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917872543989760
author Sgouritsa, Eleni
Aglietti, Virginia
Teh, Yee Whye
Doucet, Arnaud
Gretton, Arthur
Chiappa, Silvia
author_facet Sgouritsa, Eleni
Aglietti, Virginia
Teh, Yee Whye
Doucet, Arnaud
Gretton, Arthur
Chiappa, Silvia
contents The reasoning abilities of Large Language Models (LLMs) are attracting increasing attention. In this work, we focus on causal reasoning and address the task of establishing causal relationships based on correlation information, a highly challenging problem on which several LLMs have shown poor performance. We introduce a prompting strategy for this problem that breaks the original task into fixed subquestions, with each subquestion corresponding to one step of a formal causal discovery algorithm, the PC algorithm. The proposed prompting strategy, PC-SubQ, guides the LLM to follow these algorithmic steps, by sequentially prompting it with one subquestion at a time, augmenting the next subquestion's prompt with the answer to the previous one(s). We evaluate our approach on an existing causal benchmark, Corr2Cause: our experiments indicate a performance improvement across five LLMs when comparing PC-SubQ to baseline prompting strategies. Results are robust to causal query perturbations, when modifying the variable names or paraphrasing the expressions.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13952
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Prompting Strategies for Enabling Large Language Models to Infer Causation from Correlation
Sgouritsa, Eleni
Aglietti, Virginia
Teh, Yee Whye
Doucet, Arnaud
Gretton, Arthur
Chiappa, Silvia
Computation and Language
Artificial Intelligence
Machine Learning
The reasoning abilities of Large Language Models (LLMs) are attracting increasing attention. In this work, we focus on causal reasoning and address the task of establishing causal relationships based on correlation information, a highly challenging problem on which several LLMs have shown poor performance. We introduce a prompting strategy for this problem that breaks the original task into fixed subquestions, with each subquestion corresponding to one step of a formal causal discovery algorithm, the PC algorithm. The proposed prompting strategy, PC-SubQ, guides the LLM to follow these algorithmic steps, by sequentially prompting it with one subquestion at a time, augmenting the next subquestion's prompt with the answer to the previous one(s). We evaluate our approach on an existing causal benchmark, Corr2Cause: our experiments indicate a performance improvement across five LLMs when comparing PC-SubQ to baseline prompting strategies. Results are robust to causal query perturbations, when modifying the variable names or paraphrasing the expressions.
title Prompting Strategies for Enabling Large Language Models to Infer Causation from Correlation
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2412.13952