通过结构化输入让大模型更高效推理，不改模型只调输入。
Focused Chain-of-Thought: Efficient LLM Reasoning via Structured Input Information
- 用结构化信息提取替代冗长思考，聚焦关键内容。
- 算术题上减少2-3倍生成词数，准确率基本不变。
- 无需训练，适合对效率敏感的部署场景。
近期大语言模型通过生成详细的思维链（chain-of-thought）实现强大推理能力，但常导致词元消耗过多和推理延迟高。现有高效方法多聚焦模型层面干预，如强化学习或监督微调以减少冗余。本文提出一种无需训练、以输入为中心的方法：受认知心理学启发，引入聚焦思维链（Focused Chain-of-Thought, F-CoT）。F-CoT 先将查询中的关键信息提炼为简洁、结构化的上下文，再引导模型仅基于此上下文进行推理。通过避免关注无关细节，自然生成更短的推理路径。在算术应用题上，F-CoT 实现2-3倍的生成词元减少，同时保持与标准零样本思维链相当的准确率。结果表明，结构化输入是提升大模型推理效率的简单而有效手段。
原文摘要 · Abstract (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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。