Can Stories Help LLMs Reason? Curating Information Space Through Narrative

TL;DR

This paper explores how incorporating narrative structures into prompts can improve Large Language Models' ability to solve complex scientific and mathematical problems by enhancing comprehension and contextual understanding.

Contribution

It introduces the Story of Thought (SoT) method, a novel narrative-based prompting technique that outperforms traditional methods in problem-solving tasks.

Findings

SoT improves accuracy on physics, chemistry, math, and biology questions.

Narrative prompts help LLMs better understand causal relationships.

Enhanced problem comprehension through narrative structures.

Abstract

Narratives are widely recognized as a powerful tool for structuring information and facilitating comprehension of complex ideas in various domains such as science communication. This paper investigates whether incorporating narrative elements can assist Large Language Models (LLMs) in solving complex problems more effectively. We propose a novel approach, Story of Thought (SoT), integrating narrative structures into prompting techniques for problem-solving. This approach involves constructing narratives around problem statements and creating a framework to identify and organize relevant information. Our experiments show that using various LLMs with SoT consistently surpasses using them with other techniques on physics, chemistry, math, and biology questions in both the GPQA and JEEBench datasets. The narrative-based information curation process in SoT enhances problem comprehension by contextualizing critical in-domain information and highlighting causal relationships within the problem space.