让大模型学会分解任务并复用解法，提升智能体在新场景下的成功率。
Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents
- 通过自动分解任务，从成功执行中提取可复用的策略组件。
- 在新应用上达97.8%准确率，比静态方法高15.8个百分点。
- 适合需要高效适应新任务的大模型智能体研究者使用。
我们提出一种动态策略学习方法，结合广义规划与层次化任务分解，用于基于大语言模型的智能体。该方法——分层组件学习广义策略（HCL-GP）——学习能跨任务实例泛化的参数化策略，并自动从成功执行中提取可复用组件，组织成组件库以支持组合式策略生成。针对三大挑战：(1) 通过自动化分解学习组件，(2) 提升组件泛化能力以最大化复用，(3) 通过语义搜索实现高效检索。在AppWorld基准测试中，该方法在常规任务上达到98.2%准确率，在未见应用的挑战任务上达97.8%，较静态合成方法提升15.8分。对于开源模型，动态复用使成功率提升至62.5%，而无复用时接近零。结果表明，经典规划思想可有效融入大模型智能体，显著提升准确率与效率。
原文摘要 · Abstract (English)
We present a dynamic policy-learning approach that combines generalized planning and hierarchical task decomposition for LLM-based agents. Our method, Hierarchical Component Learning for Generalized Policies (HCL-GP ), learns parameterized policies that generalize across task instances and automatically extracts reusable components from successful executions, organizing them into a component library for compositional policy generation. We address three challenges: (1) learning components through automated decomposition, (2) generalizing components to maximize reuse, and (3) efficient retrieval via semantic search. Evaluated on the AppWorld benchmark, our approach achieves 98.2% accuracy on normal tasks and 97.8% on challenge tasks with unseen applications, improving 15.8 points over static synthesis on challenging scenarios. For open-source models, dynamic reuse enables 62.5% success versus near-zero without reuse. This demonstrates that classical planning concepts can be effectively integrated with LLM agents for improved accuracy and efficiency.
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。