arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.13545cs.CLcs.AI
收藏

LLMs 只有在被告知时才能进行规划

LLMs Can Plan Only If We Tell Them

Virginia Tech(弗吉尼亚理工大学)

Bilgehan Sel, Ruoxi Jia, Ming Jin

AI总结:

本文研究 LLMs 能否自主生成长时程规划,提出 AoT+ 改进方法,在规划基准上取得最先进结果,超越人类基线。

摘要:

大型语言模型(LLMs)在自然语言处理和推理方面展现了显著的能力,然而它们在自主规划方面的有效性一直备受争议。尽管现有研究利用带有外部反馈机制的 LLMs 或在受控环境中进行规划,但这些方法往往需要大量的计算和开发资源,因为需要精心设计和迭代式反向提示。此外,即使像 GPT-4 这样最先进的 LLMs,在没有额外支持的情况下,也难以在标准规划基准(如 Blocksworld)上达到人类水平的表现。本文探讨了 LLMs 是否能够独立生成与人类基线相媲美的长时程规划。我们对思维算法(Algorithm-of-Thoughts, AoT)的新颖改进(我们称之为 AoT+)帮助在规划基准上取得了最先进的结果,完全自主地超越了先前方法和人类基线。

英文摘要:

Large language models (LLMs) have demonstrated significant capabilities in natural language processing and reasoning, yet their effectiveness in autonomous planning has been under debate. While existing studies have utilized LLMs with external feedback mechanisms or in controlled environments for planning, these approaches often involve substantial computational and development resources due to the requirement for careful design and iterative backprompting. Moreover, even the most advanced LLMs like GPT-4 struggle to match human performance on standard planning benchmarks, such as the Blocksworld, without additional support. This paper investigates whether LLMs can independently generate long-horizon plans that rival human baselines. Our novel enhancements to Algorithm-of-Thoughts (AoT), which we dub AoT+, help achieve state-of-the-art results in planning benchmarks out-competing prior methods and human baselines all autonomously.

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。

↑