Chopping Trees: Semantic Similarity Based Dynamic Pruning for Tree-of-Thought Reasoning

TL;DR

This paper presents SSDP, a novel framework that dynamically prunes redundant reasoning paths in Tree-of-Thought reasoning, significantly improving efficiency while maintaining accuracy in large language model problem-solving.

Contribution

Introduces SSDP, the first online semantic merging method for real-time pruning in parallelized tree search for LLM reasoning.

Findings

Achieves up to 2.3x speedup over state-of-the-art methods

Reduces explored nodes by 85-90%

Maintains accuracy within 5% of the strongest baseline

Abstract

Tree-of-Thought (ToT) reasoning boosts the problem-solving abilities of Large Language Models (LLMs) but is computationally expensive due to semantic redundancy, where distinct branches explore equivalent reasoning paths. We introduce Semantic Similarity-Based Dynamic Pruning (SSDP), a lightweight method that, to the best of our knowledge, is the first framework to integrate online semantic merging into parallelized tree search, enabling the clustering and pruning of redundant steps in real time. Across reasoning benchmarks, including GSM8K and MATH500, SSDP achieves up to a 2.3x speedup over state-of-the-art tree-search baselines while maintaining competitive accuracy (typically within 5% of the strongest baseline) and reducing the number of explored nodes by 85-90%, demonstrating a practical approach to efficient, scalable LLM reasoning. The implementation of SSDP is publicly available at https://github.com/kimjoonghokim/SSDP.