Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sohrabi, Shirin, Ananthakrishnan, Haritha, Kokel, Harsha, Srinivas, Kavitha, Katz, Michael
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913101122633728
author Sohrabi, Shirin
Ananthakrishnan, Haritha
Kokel, Harsha
Srinivas, Kavitha
Katz, Michael
author_facet Sohrabi, Shirin
Ananthakrishnan, Haritha
Kokel, Harsha
Srinivas, Kavitha
Katz, Michael
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06957
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents
Sohrabi, Shirin
Ananthakrishnan, Haritha
Kokel, Harsha
Srinivas, Kavitha
Katz, Michael
Artificial Intelligence
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.
title Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents
topic Artificial Intelligence
url https://arxiv.org/abs/2605.06957