Eliciting Causal Abilities in Large Language Models for Reasoning Tasks

TL;DR

This paper introduces SCIE, a method to improve large language models' reasoning by eliciting their causal inference abilities through optimized prompting, leading to better performance with lower training costs.

Contribution

We propose a novel approach that enhances LLM reasoning by extracting and utilizing causal relationships from prompts, improving interpretability and reusability.

Findings

SCIE improves reasoning accuracy in LLMs.

Reduces prompt training costs significantly.

Provides interpretable causal insights for prompts.

Abstract

Prompt optimization automatically refines prompting expressions, unlocking the full potential of LLMs in downstream tasks. However, current prompt optimization methods are costly to train and lack sufficient interpretability. This paper proposes enhancing LLMs' reasoning performance by eliciting their causal inference ability from prompting instructions to correct answers. Specifically, we introduce the Self-Causal Instruction Enhancement (SCIE) method, which enables LLMs to generate high-quality, low-quantity observational data, then estimates the causal effect based on these data, and ultimately generates instructions with the optimized causal effect. In SCIE, the instructions are treated as the treatment, and textual features are used to process natural language, establishing causal relationships through treatments between instructions and downstream tasks. Additionally, we propose applying Object-Relational (OR) principles, where the uncovered causal relationships are treated as the inheritable class across task objects, ensuring low-cost reusability. Extensive experiments demonstrate that our method effectively generates instructions that enhance reasoning performance with reduced training cost of prompts, leveraging interpretable textual features to provide actionable insights.