置信度至关重要：重新审视大语言模型的内在自我纠错能力
Confidence Matters: Revisiting Intrinsic Self-Correction Capabilities of Large Language Models
AI总结：
该研究发现大语言模型自我纠错中置信度是关键潜在因素，提出IoE提示框架引导模型评估自身置信度以实现内在纠错，实验证明该方法能持续提升纠错后回答的准确率。
摘要：
大语言模型（LLMs）近期的成功引发了人们对其自我纠错能力日益增长的兴趣。本文对LLMs的内在自我纠错展开全面研究，试图解决围绕其可行性的持续争议。我们的研究发现了自我纠错过程中一个重要的潜在因素——LLMs的“置信度”。忽视这一因素可能导致模型过度自我批判，从而得出关于自我纠错有效性的不可靠结论。我们通过实验观察到，LLMs具备理解自身回答“置信度”的能力。这启发我们开发了“If-or-Else”（IoE，若-否则）提示框架，旨在引导LLMs评估自身的“置信度”，促进内在自我纠错。我们开展了大量实验，证明基于IoE的提示能让自我纠错后的回答准确率相较于初始回答获得持续提升。本研究不仅揭示了影响LLMs自我纠错的潜在因素，还提出了一个实用框架，利用IoE提示原则通过“置信度”高效提升自我纠错能力。代码可在https://github.com/MBZUAI-CLeaR/IoE-Prompting.git获取。
英文摘要：
The recent success of Large Language Models (LLMs) has catalyzed an increasing interest in their self-correction capabilities. This paper presents a comprehensive investigation into the intrinsic self-correction of LLMs, attempting to address the ongoing debate about its feasibility. Our research has identified an important latent factor - the "confidence" of LLMs - during the self-correction process. Overlooking this factor may cause the models to over-criticize themselves, resulting in unreliable conclusions regarding the efficacy of self-correction. We have experimentally observed that LLMs possess the capability to understand the "confidence" in their own responses. It motivates us to develop an "If-or-Else" (IoE) prompting framework, designed to guide LLMs in assessing their own "confidence", facilitating intrinsic self-corrections. We conduct extensive experiments and demonstrate that our IoE-based Prompt can achieve a consistent improvement regarding the accuracy of self-corrected responses over the initial answers. Our study not only sheds light on the underlying factors affecting self-correction in LLMs, but also introduces a practical framework that utilizes the IoE prompting principle to efficiently improve self-correction capabilities with "confidence". The code is available at https://github.com/MBZUAI-CLeaR/IoE-Prompting.git.
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。
