arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.11698cs.CL
收藏

面向大型语言模型的元推理

Meta Reasoning for Large Language Models

Tsinghua University(清华大学);Microsoft Research(微软研究院)

Peizhong Gao, Ao Xie, Shaoguang Mao, Wenshan Wu, Yan Xia, Haipeng Mi, Furu Wei

AI总结:

本文提出 MRP 元推理提示方法,通过两阶段动态选择并应用合适推理策略,使 LLM 在多类任务上达到或接近 SOTA,并提升计算效率与适应性。

摘要:

我们提出了 Meta-Reasoning Prompting(MRP,元推理提示),这是一种受人类元推理启发、面向大型语言模型(LLM)的新颖且高效的系统提示方法。传统的基于上下文学习的推理技术(例如 Tree-of-Thoughts)虽展现出良好前景,但由于其专用性质,无法在多样化任务上持续取得最先进性能。MRP 通过引导 LLM 根据每项任务的具体要求动态选择并应用不同推理方法来解决这一局限,从而同时优化性能与计算效率。借助 MRP,LLM 推理分为两个阶段运行。首先,LLM 利用任务输入线索以及可用方法的客观描述,识别最合适的推理方法。随后,它应用所选方法完成任务。这种动态策略模拟了人类的元推理,使模型能够在广泛的问题领域中表现出色。我们通过全面的基准测试评估 MRP 的有效性。结果表明,MRP 在多样化任务上达到或接近最先进性能。MRP 代表了一项重要进展,使 LLM 能够识别不同问题中的认知挑战,并利用不同推理方法的优势,增强其高效处理多样且复杂问题领域的能力。每个 LLM 都值得拥有 Meta-Reasoning Prompting,以释放其全部潜力,并确保在不断演变的挑战与应用环境中具备适应性。

英文摘要:

We introduce Meta-Reasoning Prompting (MRP), a novel and efficient system prompting method for large language models (LLMs) inspired by human meta-reasoning. Traditional in-context learning-based reasoning techniques, such as Tree-of-Thoughts, show promise but lack consistent state-of-the-art performance across diverse tasks due to their specialized nature. MRP addresses this limitation by guiding LLMs to dynamically select and apply different reasoning methods based on the specific requirements of each task, optimizing both performance and computational efficiency. With MRP, LLM reasoning operates in two phases. Initially, the LLM identifies the most appropriate reasoning method using task input cues and objective descriptions of available methods. Subsequently, it applies the chosen method to complete the task. This dynamic strategy mirrors human meta-reasoning, allowing the model to excel in a wide range of problem domains. We evaluate the effectiveness of MRP through comprehensive benchmarks. The results demonstrate that MRP achieves or approaches state-of-the-art performance across diverse tasks. MRP represents a significant advancement in enabling LLMs to identify cognitive challenges across problems and leverage benefits across different reasoning approaches, enhancing their ability to handle diverse and complex problem domains efficiently. Every LLM deserves a Meta-Reasoning Prompting to unlock its full potential and ensure adaptability in an ever-evolving landscape of challenges and applications.

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。

↑