Exploring the Role of Reasoning Structures for Constructing Proofs in Multi-Step Natural Language Reasoning with Large Language Models

TL;DR

This paper investigates how large language models can better construct multi-step proofs by leveraging reasoning structures through in-context learning, demonstrating that structure-aware techniques enhance performance and explainability.

Contribution

The study introduces structure-aware demonstration and pruning methods, showing they improve proof construction in large language models during multi-step reasoning tasks.

Findings

Structure-aware methods improve proof accuracy.

Enhanced explainability of reasoning processes.

Performance gains observed across multiple tasks.

Abstract

When performing complex multi-step reasoning tasks, the ability of Large Language Models (LLMs) to derive structured intermediate proof steps is important for ensuring that the models truly perform the desired reasoning and for improving models' explainability. This paper is centred around a focused study: whether the current state-of-the-art generalist LLMs can leverage the structures in a few examples to better construct the proof structures with \textit{in-context learning}. Our study specifically focuses on structure-aware demonstration and structure-aware pruning. We demonstrate that they both help improve performance. A detailed analysis is provided to help understand the results.