SelfCheck：利用大语言模型零样本检查其自身的逐步推理过程
SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning
AI总结：
针对大语言模型处理复杂非线性问题时推理易出错的问题，提出通用零样本验证框架SelfCheck，可让模型自查推理错误，结合多解答加权投票提升了三个数学数据集上的问答准确率。
摘要：
大语言模型（LLM）近期取得的进展，尤其是思维链提示技术的提出，使得通过逐步推理自动回答问题成为可能。但在面对需要非线性思维的更复杂问题时，即便性能最强的LLM也会出错。为解决这一问题，我们探究了LLM无需借助外部资源，能否识别自身逐步推理过程中的错误。为此，我们提出SelfCheck——一种用于识别此类错误的通用零样本验证框架。随后，我们利用这些检查结果，通过对问题的多个解答进行加权投票，来提升问答性能。我们在GSM8K、MathQA和MATH三个数据集上测试了SelfCheck，发现它能成功识别错误，进而提升最终答案准确率。
英文摘要：
The recent progress in large language models (LLMs), especially the invention of chain-of-thought prompting, has made it possible to automatically answer questions by stepwise reasoning. However, when faced with more complicated problems that require non-linear thinking, even the strongest LLMs make mistakes. To address this, we explore whether LLMs are able to recognize errors in their own step-by-step reasoning, without resorting to external resources. To this end, we propose SelfCheck, a general-purpose zero-shot verification schema for recognizing such errors. We then use the results of these checks to improve question-answering performance by conducting weighted voting on multiple solutions to the question. We test SelfCheck on three datasets (GSM8K, MathQA, and MATH) and find that it successfully recognizes errors and, in turn, increases final answer accuracies.
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。
