Tree of Problems: Improving structured problem solving with compositionality

TL;DR

This paper introduces Tree of Problems (ToP), a simplified problem-solving framework that enhances complex reasoning in large language models by leveraging task compositionality, outperforming existing methods like ToT, GoT, and CoT.

Contribution

The paper proposes Tree of Problems (ToP), a new approach that improves complex reasoning by dividing tasks into identical subtasks, demonstrating superior performance over existing methods.

Findings

ToP outperforms ToT and GoT on complex tasks.

ToP surpasses CoT in complex reasoning tasks.

Empirical results validate ToP's effectiveness.

Abstract

Large Language Models (LLMs) have demonstrated remarkable performance across multiple tasks through in-context learning. For complex reasoning tasks that require step-by-step thinking, Chain-of-Thought (CoT) prompting has given impressive results, especially when combined with self-consistency. Nonetheless, some tasks remain particularly difficult for LLMs to solve. Tree of Thoughts (ToT) and Graph of Thoughts (GoT) emerged as alternatives, dividing the complex problem into paths of subproblems. In this paper, we propose Tree of Problems (ToP), a simpler version of ToT, which we hypothesise can work better for complex tasks that can be divided into identical subtasks. Our empirical results show that our approach outperforms ToT and GoT, and in addition performs better than CoT on complex reasoning tasks. All code for this paper is publicly available here: https://github.com/ArmelRandy/tree-of-problems.