Saved in:
Bibliographic Details
Main Authors: Ling, Guoming, Huang, Zhongzhan, Lin, Yupei, Li, Junxin, Zhong, Shanshan, Wu, Hefeng, Lin, Liang
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.11340
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914473091006464
author Ling, Guoming
Huang, Zhongzhan
Lin, Yupei
Li, Junxin
Zhong, Shanshan
Wu, Hefeng
Lin, Liang
author_facet Ling, Guoming
Huang, Zhongzhan
Lin, Yupei
Li, Junxin
Zhong, Shanshan
Wu, Hefeng
Lin, Liang
contents Chain-of-Thought reasoning has significantly enhanced the problem-solving capabilities of Large Language Models. Unfortunately, current models generate reasoning steps sequentially without foresight, often becoming trapped in suboptimal reasoning paths with redundant steps. In contrast, we introduce Neural Chain-of-Thought Search (NCoTS), a framework that reformulates reasoning as a dynamic search for the optimal thinking strategy. By quantitatively characterizing the solution space, we reveal the existence of sparse superior reasoning paths that are simultaneously more accurate and concise than standard outputs. Our method actively navigates towards these paths by evaluating candidate reasoning operators using a dual-factor heuristic that optimizes for both correctness and computational cost. Consequently, NCoTS achieves a Pareto improvement across diverse reasoning benchmarks, boosting accuracy by over 3.5% while reducing generation length by over 22%. Our code and data are available at https://github.com/MilkThink-Lab/Neural-CoT-Search.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11340
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models
Ling, Guoming
Huang, Zhongzhan
Lin, Yupei
Li, Junxin
Zhong, Shanshan
Wu, Hefeng
Lin, Liang
Computation and Language
Chain-of-Thought reasoning has significantly enhanced the problem-solving capabilities of Large Language Models. Unfortunately, current models generate reasoning steps sequentially without foresight, often becoming trapped in suboptimal reasoning paths with redundant steps. In contrast, we introduce Neural Chain-of-Thought Search (NCoTS), a framework that reformulates reasoning as a dynamic search for the optimal thinking strategy. By quantitatively characterizing the solution space, we reveal the existence of sparse superior reasoning paths that are simultaneously more accurate and concise than standard outputs. Our method actively navigates towards these paths by evaluating candidate reasoning operators using a dual-factor heuristic that optimizes for both correctness and computational cost. Consequently, NCoTS achieves a Pareto improvement across diverse reasoning benchmarks, boosting accuracy by over 3.5% while reducing generation length by over 22%. Our code and data are available at https://github.com/MilkThink-Lab/Neural-CoT-Search.
title Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2601.11340