理解先于推理：通过迭代摘要预提示增强思维链
Understanding Before Reasoning: Enhancing Chain-of-Thought with Iterative Summarization Pre-Prompting
AI总结：
针对思维链在关键信息缺失时推理困难的问题，提出迭代摘要预提示方法ISP^2，通过提取、评估并合并信息对，最终将唯一关键信息与问题输入大语言模型，提升推理性能7.1%。
摘要：
思维链（Chain-of-Thought, CoT）提示是大语言模型（LLMs）中增强复杂推理的主流范式。它引导大语言模型呈现多步推理，而非直接生成最终答案。然而，当推理所需的关键信息是隐含的或缺失时，CoT会遇到困难。这是因为CoT强调推理步骤的顺序，而忽略了早期提取关键信息。我们提出了一种名为迭代摘要预提示（Iterative Summarization Pre-Prompting, ISP^2）的预提示方法，以在关键信息未明确提供时优化大语言模型的推理。首先，提取实体及其对应描述，形成潜在的关键信息对。接下来，我们使用可靠性评分来评估这些信息对，然后将评分最低的两对合并为一个新的实体描述。重复此过程，直到获得唯一的关键信息对。最后，将该信息对与原始问题一起输入大语言模型以生成答案。大量实验表明，与现有方法相比，该方法提升了7.1%。与传统提示不同，ISP^2采用归纳式预提示方法，可灵活集成到多种推理框架中。代码可在https://github.com/zdhgreat/ISP-2获取。
英文摘要：
Chain-of-Thought (CoT) Prompting is a dominant paradigm in Large Language Models (LLMs) to enhance complex reasoning. It guides LLMs to present multi-step reasoning, rather than generating the final answer directly. However, CoT encounters difficulties when key information required for reasoning is implicit or missing. This occurs because CoT emphasizes the sequence of reasoning steps while overlooking the early extraction of essential information. We propose a pre-prompting method called Iterative Summarization Pre-Prompting (ISP^2) to refine LLM reasoning when key information is not explicitly provided. First, entities and their corresponding descriptions are extracted to form potential key information pairs. Next, we use a reliability rating to assess these pairs, then merge the two lowest-ranked pairs into a new entity description. This process is repeated until a unique key information pair is obtained. Finally, that pair, along with the original question, is fed into LLMs to produce the answer. Extensive experiments demonstrate a 7.1% improvement compared to existing methods. Unlike traditional prompting, ISP^2 adopts an inductive approach with pre-prompting, offering flexible integration into diverse reasoning frameworks. The code is available at https://github.com/zdhgreat/ISP-2.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
