The performance of Large Language Models (LLMs) in reasoning tasks depends heavily on prompt design, with Chain-of-Thought (CoT) and self-consistency being critical methods that enhance this ability. However, these methods do not fully exploit the answers generated by the LLM to guide subsequent responses. This paper proposes a new prompting method, named Progressive-Hint Prompting (PHP), that enables automatic multiple interactions between users and LLMs by using previously generated answers as hints to progressively guide toward the correct answers. PHP is orthogonal to CoT and self-consistency, making it easy to combine with state-of-the-art techniques to further improve performance. We conducted an extensive and comprehensive evaluation to demonstrate the effectiveness of the proposed method. Our experimental results on six benchmarks show that combining CoT and self-consistency with PHP significantly improves accuracy while remaining highly efficient. For instance, with text-davinci-003, we observed a 4.2% improvement on GSM8K with greedy decoding compared to Complex CoT, and a 46.17% reduction in sample paths with self-consistency. With GPT-4 and PHP, we achieve state-of-the-art performances on SVAMP (89.1% -> 91.9%), GSM8K (92% -> 95.5%), AQuA (76.4% -> 79.9%) and MATH (50.2% -> 53.9%).

翻译：大语言模型（LLMs）在推理任务中的表现高度依赖于提示设计，其中思维链（CoT）和自一致性是增强该能力的关键方法。然而，这些方法并未充分利用LLM生成的答案来引导后续响应。本文提出一种名为渐进式提示（Progressive-Hint Prompting, PHP）的新型提示方法，通过将先前生成的答案作为提示，逐步引导用户与LLM之间实现自动化的多轮交互，最终趋向正确解答。PHP与CoT和自一致性正交，可轻松与前沿技术结合以进一步提升性能。我们进行了广泛而全面的评估以证明该方法的有效性。在六个基准测试上的实验结果表明，将CoT和自一致性与PHP结合，在保持高效性的同时显著提升了准确率。例如，使用text-davinci-003模型时，与复杂CoT相比，贪心解码在GSM8K上实现了4.2%的提升；在自一致性条件下，样本路径减少了46.17%。采用GPT-4与PHP，我们在SVAMP（89.1% -> 91.9%）、GSM8K（92% -> 95.5%）、AQuA（76.4% -> 79.9%）和MATH（50.2% -> 53.9%）上达到了当前最优性能。