aiqbench analysis
This paper proposes using Large Language Models (LLMs) to automatically generate instance generators for PDDL planning domains, addressing the scalability limitations of prior reasoning and learning-based methods. By incorporating built-in soundness checks, the approach efficiently produces large numbers of valid and diverse planning instances without relying on manual coding. This advancement directly enhances the automation capabilities within the AI Planning community by bridging formal logic specifications with generative model flexibility.
Generating Instance Generators in PDDL Planning
arXiv:2609.06071v1 Announce Type: new Abstract: PDDL, the de-facto standard language in the AI Planning community, is designed to specify planning domains: sets of instances that share the same predicates and action schemas. Yet it does not provide any means to specify the actual instance set, i.e., legality constraints on initial states and goal conditions, as well as possibly domain subset constraints specifying an instance subset we are interested in. One consequence of this is that instance generation has always been ad-hoc, with manually written domain- and subset-specific instance generators. Recent work has started to address this, through reasoning and learning methods that however suffer from scalability limitations. Here we introduce an alternative approach, leveraging LLMs to generate instance-generation programs, with built-in soundness guarantees through prescribed checks. We show that these automatically generated instance generators return large numbers of sound and diverse instances efficiently.
Abstract:PDDL, the de-facto standard language in the AI Planning community, is designed to specify planning domains: sets of instances that share the same predicates and action schemas. Yet it does not provide any means to specify the actual instance set, i.e., legality constraints on initial states and goal conditions, as well as possibly domain subset constraints specifying an instance subset we are interested in. One consequence of this is that instance generation has always been ad-hoc, with manually written domain- and subset-specific instance generators. Recent work has started to address this, through reasoning and learning methods that however suffer from scalability limitations. Here we introduce an alternative approach, leveraging LLMs to generate instance-generation programs, with built-in soundness guarantees through prescribed checks. We show that these automatically generated instance generators return large numbers of sound and diverse instances efficiently.
Artificial Intelligence (cs.AI)
arXiv:2609.06071 [cs.AI]
(or arXiv:2609.06071v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.06071 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 232 words
Related coverage
Provably Complete Generalized Planning with LLMs
arXiv · Sep 24, 2026
Grounded Evaluation and Repair for NL-to-PDDL Problem Generation
arXiv · Sep 11, 2026
Neurosymbolic Embodied Agents
arXiv · Aug 18, 2026
A Temporal Planning Framework for Disruption Aware Dynamic Route Optimization in Heterogeneous Railway Systems
arXiv · Jun 15, 2026