用大模型分析PDDL文本，提前剔除无关内容，大幅加速规划求解
Semantic Partial Grounding via LLMs
- 利用大模型解析PDDL文件，预判并移除冗余对象与操作
- 在7个难题上实现数倍至数十倍的接地速度提升
- 适合需要快速规划且对精度要求不苛刻的系统
在经典规划中，接地是关键步骤，但随任务规模扩大，其生成的可执行动作和原子项呈指数增长，造成计算瓶颈。近期部分接地方法通过预测模型仅逐步接地高潜力算子缓解此问题。然而，这些方法主要依赖关系特征或学习嵌入，未充分使用PDDL描述中的文本与结构线索。本文提出SPG-LLM，利用大语言模型分析领域与问题文件，启发式识别可能无关的对象、动作与谓词，从而显著缩小接地任务规模。在七个难以接地的基准测试中，SPG-LLM实现了更快的接地速度——有时快达数量级，并在部分领域取得相当或更优的计划代价。
原文摘要 · Abstract (English)
Grounding is a critical step in classical planning, yet it often becomes a computational bottleneck due to the exponential growth in grounded actions and atoms as task size increases. Recent advances in partial grounding have addressed this challenge by incrementally grounding only the most promising operators, guided by predictive models. However, these approaches primarily rely on relational features or learned embeddings and do not leverage the textual and structural cues present in PDDL descriptions. We propose SPG-LLM, which uses LLMs to analyze the domain and problem files to heuristically identify potentially irrelevant objects, actions, and predicates prior to grounding, significantly reducing the size of the grounded task. Across seven hard-to-ground benchmarks, SPG-LLM achieves faster grounding-often by orders of magnitude-while delivering comparable or better plan costs in some domains.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。