LPML: LLM-Prompting Markup Language for Mathematical Reasoning

TL;DR

This paper introduces LPML, a markup language that enables large language models to improve mathematical reasoning by integrating structured prompts, external Python computation, and chain-of-thought methods, enhancing accuracy in problem-solving.

Contribution

The paper presents a novel framework that combines a markup language with external Python tools and chain-of-thought prompting to improve LLMs' mathematical reasoning capabilities.

Findings

LPML improves reasoning accuracy in LLMs.

Combining CoT with external tools reduces errors.

Zero-shot prompting achieves effective mathematical reasoning.

Abstract

In utilizing large language models (LLMs) for mathematical reasoning, addressing the errors in the reasoning and calculation present in the generated text by LLMs is a crucial challenge. In this paper, we propose a novel framework that integrates the Chain-of-Thought (CoT) method with an external tool (Python REPL). We discovered that by prompting LLMs to generate structured text in XML-like markup language, we could seamlessly integrate CoT and the external tool and control the undesired behaviors of LLMs. With our approach, LLMs can utilize Python computation to rectify errors within CoT. We applied our method to ChatGPT (GPT-3.5) to solve challenging mathematical problems and demonstrated that combining CoT and Python REPL through the markup language enhances the reasoning capability of LLMs. Our approach enables LLMs to write the markup language and perform advanced mathematical reasoning using only zero-shot prompting.