Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


L3M+P: Lifelong Planning with Large Language Models

L3M+P: Lifelong Planning with Large Language Models.
Krish Agarwal, Yuqian Jiang, Jiaheng Hu, Bo Liu, and Peter Stone.
In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), October 2025.

Download

[PDF]1.2MB  

Abstract

By combining classical planning methods with large language models (LLMs), recentresearch such as LLM+P has enabled agents to plan for general tasks given innatural language. However, scaling these methods to general-purpose servicerobots remains challenging: (1) classical planning algorithms generally require adetailed and consistent specification of the environment, which is not alwaysreadily available; and (2) existing frameworks mainly focus on isolated planningtasks, whereas robots are often meant to serve in long-term continuousdeployments, and therefore must maintain a dynamic memory of the environmentwhich can be updated with multi-modal inputs and extracted as planning knowledgefor future tasks. To address these two issues, this paper introduces L3M+P(Lifelong LLM+P), a framework that uses an external knowledge graph as arepresentation of the world state. The graph can be updated from multiple sourcesof information, including sensory input and natural language interactions withhumans. L3M+P enforces rules for the expected format of the absolute world stategraph to maintain consistency between graph updates. At planning time, given anatural language description of a task, L3M+P retrieves context from theknowledge graph and generates a problem definition for classical planners.Evaluated on household robot simulators and on a real-world service robot, L3M+Pachieves significant improvement over baseline methods both on accuratelyregistering natural language state changes and on correctly generating plans,thanks to the knowledge graph retrieval and verification.

BibTeX Entry

@InProceedings{agarwal2025_l3mp,
  author   = {Krish Agarwal and Yuqian Jiang and Jiaheng Hu and Bo Liu and Peter Stone},
  title    = {L3M+P: Lifelong Planning with Large Language Models},
  booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year     = {2025},
  month    = {October},
  location = {Hangzhou, China},
  abstract = {By combining classical planning methods with large language models (LLMs), recent
research such as LLM+P has enabled agents to plan for general tasks given in
natural language. However, scaling these methods to general-purpose service
robots remains challenging: (1) classical planning algorithms generally require a
detailed and consistent specification of the environment, which is not always
readily available; and (2) existing frameworks mainly focus on isolated planning
tasks, whereas robots are often meant to serve in long-term continuous
deployments, and therefore must maintain a dynamic memory of the environment
which can be updated with multi-modal inputs and extracted as planning knowledge
for future tasks. To address these two issues, this paper introduces L3M+P
(Lifelong LLM+P), a framework that uses an external knowledge graph as a
representation of the world state. The graph can be updated from multiple sources
of information, including sensory input and natural language interactions with
humans. L3M+P enforces rules for the expected format of the absolute world state
graph to maintain consistency between graph updates. At planning time, given a
natural language description of a task, L3M+P retrieves context from the
knowledge graph and generates a problem definition for classical planners.
Evaluated on household robot simulators and on a real-world service robot, L3M+P
achieves significant improvement over baseline methods both on accurately
registering natural language state changes and on correctly generating plans,
thanks to the knowledge graph retrieval and verification.
  },
}

Generated by bib2html.pl (written by Patrick Riley ) on Mon Sep 29, 2025 19:22:36