Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs

TL;DR

This paper introduces Chain of Preference Optimization (CPO), a fine-tuning method that enhances chain-of-thought reasoning in large language models by aligning them with the more optimal reasoning paths found through tree-of-thought search, achieving better performance without increased inference complexity.

Contribution

The paper proposes CPO, a novel fine-tuning approach that leverages tree-of-thought search paths to improve chain-of-thought reasoning in LLMs, reducing inference costs.

Findings

CPO improves LLM performance on complex reasoning tasks.

CPO achieves comparable or better results than tree-of-thought search.

CPO reduces inference complexity while enhancing reasoning quality.

Abstract

The recent development of chain-of-thought (CoT) decoding has enabled large language models (LLMs) to generate explicit logical reasoning paths for complex problem-solving. However, research indicates that these paths are not always deliberate and optimal. The tree-of-thought (ToT) method employs tree-searching to extensively explore the reasoning space and find better reasoning paths that CoT decoding might overlook. This deliberation, however, comes at the cost of significantly increased inference complexity. In this work, we demonstrate that fine-tuning LLMs leveraging the search tree constructed by ToT allows CoT to achieve similar or better performance, thereby avoiding the substantial inference burden. This is achieved through Chain of Preference Optimization (CPO), where LLMs are fine-tuned to align each step of the CoT reasoning paths with those of ToT using the inherent preference information in the tree-search process. Extensive experimental results show that CPO significantly improves LLM performance in solving a variety of complex problems, including question answering, fact verification, and arithmetic reasoning, demonstrating its effectiveness. Our code is available at https://github.com/sail-sg/CPO.