重读问题可提升大型语言模型的推理能力
Re-Reading Improves Reasoning in Large Language Models
AI总结：
本文提出 Re2 提示方法，通过让 LLM 重读输入问题增强理解并模拟双向编码，在 14 个数据集上验证其能稳定提升推理性能且兼容 CoT 等方法。
摘要：
为了增强现成大型语言模型（LLMs）的推理能力，我们提出了一种简单、通用且有效的提示方法 Re2，即对作为输入的问题进行重读（Re-Reading）。与大多数旨在从输出中激发推理过程的思维启发式提示方法（如思维链 CoT）不同，Re2 将重点转移到输入侧，通过对问题处理两次来增强理解过程。因此，Re2 展现出很强的通用性，并且与包括 CoT 在内的大多数思维启发式提示方法具有良好的兼容性。关键的是，Re2 在单向仅解码器 LLMs 中促进了“双向”编码，因为第一遍处理可以为第二遍处理提供全局信息。我们首先进行了一项初步实证研究作为 Re2 的基础，展示其实现“双向”注意力机制的潜力。随后，我们在涵盖 14 个数据集、共 112 组实验的广泛推理基准上评估 Re2，以验证其有效性和通用性。研究结果表明，除原始 ChatGPT 上的少数场景外，Re2 通过简单的重读策略能够一致地提升 LLMs 的推理性能。进一步分析揭示了 Re2 的适应性，展示其如何与不同 LLMs、思维启发式提示以及集成策略有效结合。我们的代码可在 https://github.com/Tebmer/Rereading-LLM-Reasoning/ 获取。
英文摘要：
To enhance the reasoning capabilities of off-the-shelf Large Language Models (LLMs), we introduce a simple, yet general and effective prompting method, Re2, i.e., \textbf{Re}-\textbf{Re}ading the question as input. Unlike most thought-eliciting prompting methods, such as Chain-of-Thought (CoT), which aim to elicit the reasoning process in the output, Re2 shifts the focus to the input by processing questions twice, thereby enhancing the understanding process. Consequently, Re2 demonstrates strong generality and compatibility with most thought-eliciting prompting methods, including CoT. Crucially, Re2 facilitates a "bidirectional" encoding in unidirectional decoder-only LLMs because the first pass could provide global information for the second pass. We begin with a preliminary empirical study as the foundation of Re2, illustrating its potential to enable "bidirectional" attention mechanisms. We then evaluate Re2 on extensive reasoning benchmarks across 14 datasets, spanning 112 experiments, to validate its effectiveness and generality. Our findings indicate that, with the exception of a few scenarios on vanilla ChatGPT, Re2 consistently enhances the reasoning performance of LLMs through a simple re-reading strategy. Further analyses reveal Re2's adaptability, showing how it can be effectively integrated with different LLMs, thought-eliciting prompting, and ensemble strategies. Our code is available at \url{https://github.com/Tebmer/Rereading-LLM-Reasoning/}
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。
