Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models
This addresses the issue of inefficient and inaccurate reasoning in LLMs for users relying on complex problem-solving, though it appears incremental as it builds on existing Chain-of-Thought methods.
The paper tackles the problem of suboptimal reasoning paths in Chain-of-Thought reasoning for Large Language Models by introducing Neural Chain-of-Thought Search (NCoTS), which searches for optimal reasoning strategies. The result is a Pareto improvement with accuracy boosted by over 3.5% and generation length reduced by over 22% across diverse benchmarks.
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.