Search papers, labs, and topics across Lattice.
The paper introduces Hierarchical Chain-of-Thought (Hi-CoT) prompting, a structured approach that decomposes LLM reasoning into hierarchical planning and execution substeps. This method aims to address limitations of flat CoT prompting, such as redundancy and suboptimal performance in complex, multi-step reasoning tasks. Experiments across various LLMs and mathematical reasoning benchmarks demonstrate that Hi-CoT improves average accuracy by 6.2% and reduces reasoning trace length by 13.9% compared to standard CoT.
LLMs reason more accurately and efficiently when forced to plan before acting, achieving up to 61.4% accuracy gains on certain tasks.
Chain-of-Thought (CoT) prompting has significantly improved the reasoning capabilities of large language models (LLMs). However, conventional CoT often relies on unstructured, flat reasoning chains that suffer from redundancy and suboptimal performance. In this work, we introduce Hierarchical Chain-of-Thought (Hi-CoT) prompting, a structured reasoning paradigm specifically designed to address the challenges of complex, multi-step reasoning. Hi-CoT decomposes the reasoning process into hierarchical substeps by alternating between instructional planning and step-by-step execution. This decomposition enables LLMs to better manage long reasoning horizons and maintain logical coherence. Extensive evaluations across diverse LLMs and mathematical reasoning benchmarks show that Hi-CoT consistently improves average accuracy by 6.2% (up to 61.4% on certain models and tasks) while reducing reasoning trace length by 13.9% compared to CoT prompting. We further show that accuracy and efficiency are maximized when models strictly adhere to the hierarchical structure. Our code is available at https://github.com/XingshuaiHuang/Hi-CoT.