arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2501.18784cs.AI
收藏

基于大语言模型生成启发式策略的后继生成器规划

Successor-Generator Planning with LLM-generated Heuristics

Computer Science Department, Bar-Ilan University(巴伊兰大学计算机科学系)

Alexander Tuisov, Yonatan Vernik, Alexander Shleyfman

AI总结:

研究借助大语言模型,从规划任务中生成特定问题启发式函数并集成到搜索算法。该方法在多规划基准测试中性能优异,还能解决传统形式难表达的问题,经评估展现出有效性。

摘要:

启发式策略是确定性规划的核心组成部分,尤其是在与领域无关的设置中,通用性优先于特定任务调整。鉴于大语言模型(LLMs)的最新进展,本研究重新审视该范式,其能够直接从问题定义中自动合成启发式策略,无需手工制作领域知识。我们提出一种方法,利用大语言模型从通过后继生成器、目标测试和用通用编程语言编写的初始状态指定的规划任务中生成特定问题的启发式函数。这些启发式策略被编译并集成到标准启发式搜索算法中,如贪婪最佳优先搜索。我们的方法在广泛的既定规划基准上取得了有竞争力的,且在许多情况下是最先进的性能。此外,它能够解决难以用传统形式表示的问题,包括那些具有复杂数值约束或自定义转换动态的问题。我们提供了广泛的实证评估,以表征该方法在不同规划设置中的优势和局限性,证明了其有效性。

英文摘要:

Heuristics are a central component of deterministic planning, particularly in domain-independent settings where general applicability is prioritized over task-specific tuning. This work revisits that paradigm in light of recent advances in large language models (LLMs), which enable the automatic synthesis of heuristics directly from problem definitions -- bypassing the need for handcrafted domain knowledge. We present a method that employs LLMs to generate problem-specific heuristic functions from planning tasks specified through successor generators, goal tests, and initial states written in a general-purpose programming language. These heuristics are compiled and integrated into standard heuristic search algorithms, such as greedy best-first search. Our approach achieves competitive, and in many cases state-of-the-art, performance across a broad range of established planning benchmarks. Moreover, it enables the solution of problems that are difficult to express in traditional formalisms, including those with complex numeric constraints or custom transition dynamics. We provide an extensive empirical evaluation that characterizes the strengths and limitations of the approach across diverse planning settings, demonstrating its effectiveness.

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。

↑