arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.19793cs.CL
收藏

PDDLEGO:文本环境中的迭代规划

PDDLEGO: Iterative Planning in Textual Environments

University of Pennsylvania(宾夕法尼亚大学);The University of Arizona(亚利桑那大学);Allen Institute for Artificial Intelligence(艾伦人工智能研究所)

Li Zhang, Peter Jansen, Tianyi Zhang, Peter Clark, Chris Callison-Burch, Niket Tandon

AI总结:

针对部分可观测文本环境,本文提出 PDDLEGO,通过迭代构建 PDDL 表示、执行子目标并扩充信息完成最终规划,在两个模拟基准上显著优于端到端 LLM。

摘要:

事实证明,即使对当前模型而言,在文本环境中进行规划也是一项长期挑战。近期一条很有前景的研究路线使用 LLM 生成环境的形式化表示,并由符号规划器求解。然而,现有方法依赖完全可观测的环境,即所有实体状态在初始时均已知,因此可以一次性构建表示并得到完整规划。相比之下,我们处理部分可观测环境:初始阶段没有足够信息来为最终目标制定规划。我们提出 PDDLEGO,它迭代构建规划表示,从而为给定子目标生成部分规划。通过完成该子目标,系统获得更多信息以扩充表示,最终实现最终目标。我们表明,在 Coin Collector 模拟中,少样本 PDDLEGO 生成的规划比端到端生成规划的效率高 43%;在更复杂的 Cooking World 模拟中也取得 98% 的强劲表现,而端到端 LLM 无法生成连贯规划,仅为 4%。

英文摘要:

Planning in textual environments have been shown to be a long-standing challenge even for current models. A recent, promising line of work uses LLMs to generate a formal representation of the environment that can be solved by a symbolic planner. However, existing methods rely on a fully-observed environment where all entity states are initially known, so a one-off representation can be constructed, leading to a complete plan. In contrast, we tackle partially-observed environments where there is initially no sufficient information to plan for the end-goal. We propose PDDLEGO that iteratively construct a planning representation that can lead to a partial plan for a given sub-goal. By accomplishing the sub-goal, more information is acquired to augment the representation, eventually achieving the end-goal. We show that plans produced by few-shot PDDLEGO are 43% more efficient than generating plans end-to-end on the Coin Collector simulation, with strong performance (98%) on the more complex Cooking World simulation where end-to-end LLMs fail to generate coherent plans (4%).

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。

↑