Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs
This addresses the need for more efficient and effective reasoning in LLMs for complex tasks like question answering and arithmetic reasoning, representing an incremental improvement over existing methods.
The paper tackles the problem of suboptimal reasoning paths in chain-of-thought (CoT) decoding for LLMs by fine-tuning models using tree-of-thought (ToT) search trees, achieving similar or better performance without the high inference cost of ToT.
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.