聚焦思维链：通过结构化输入信息实现高效大语言模型推理

摘要

近期，大型语言模型通过生成详尽思维链轨迹实现了强劲的推理性能，但这往往导致令牌使用量激增和推理延迟升高。现有效率优化方法通常聚焦于模型侧干预（如强化学习或监督微调）以降低冗余度。与之相反，我们提出一种免训练、输入侧驱动的创新方案。受认知心理学启发，我们引入聚焦式思维链（F-CoT）方法，将信息提取与推理过程分离。F-CoT首先将查询中的关键信息组织为简洁的结构化上下文，随后引导模型仅基于该上下文进行推理。通过避免关注无关细节，F-CoT自然生成更简短的推理路径。在数学应用题测试中，F-CoT在保持与标准零样本思维链相当准确率的同时，将生成令牌量减少2-3倍。这些结果表明，结构化输入是实现更高效大语言模型推理的简单而有效的关键抓手。

English

Recent large language models achieve strong reasoning performance by generating detailed chain-of-thought traces, but this often leads to excessive token use and high inference latency. Existing efficiency approaches typically focus on model-centric interventions, such as reinforcement learning or supervised fine-tuning, to reduce verbosity. In contrast, we propose a training-free, input-centric approach. Inspired by cognitive psychology, we introduce Focused Chain-of-Thought (F-CoT), which separates information extraction from the reasoning process. F-CoT first organizes the essential information from a query into a concise, structured context and then guides the model to reason exclusively over this context. By preventing attention to irrelevant details, F-CoT naturally produces shorter reasoning paths. On arithmetic word problems, F-CoT reduces generated tokens by 2-3x while maintaining accuracy comparable to standard zero-shot CoT. These results highlight structured input as a simple yet effective lever for more efficient LLM reasoning.

聚焦思维链：通过结构化输入信息实现高效大语言模型推理

Focused Chain-of-Thought: Efficient LLM Reasoning via Structured Input Information

摘要

Support