EUVIMEDEUVIMED entdecken
Crossref

Successor-Generator Planning with LLM-generated Heuristics

Alexander Tuisov, Yonatan Vernik, Alexander Shleyfman

Proceedings of the International Conference on Automated Planning and Scheduling · 08.06.2026 · 36 (1) · 332-341

DOI: 10.1609/icaps.v36i1.42843 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the International Conference on Automated Planning and Scheduling, online erschienen am 08.06.2026. Verfasst von Alexander Tuisov und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Heuristics are a central component of deterministic planning, particularly in domain-independent settings where general applicability is prioritized over task-specific tuning. This work revisits that paradigm in light of recent advances in large language models (LLMs), which enable the automatic synthesis of heuristics directly from problem definitions -- bypassing the need for handcrafted domain knowledge. We present a method that employs LLMs to generate problem-specific heuristic functions from planning tasks specified through successor generators, goal tests, and initial states written in a general-purpose programming language. These heuristics are compiled and integrated into standard heuristic search algorithms, such as greedy best-first search. Our approach achieves competitive, and in many cases state-of-the-art, performance across a broad range of established planning benchmarks. Moreover, it enables the solution of problems that are difficult to express in traditional formalisms, including those with complex numeric constraints or custom transition dynamics. We provide an extensive empirical evaluation that characterizes the strengths and limitations of the approach across diverse planning settings, demonstrating its effectiveness.

Zitieren

APA

Tuisov, A., Vernik, Y., & Shleyfman, A. (2026). Successor-Generator Planning with LLM-generated Heuristics. Proceedings of the International Conference on Automated Planning and Scheduling, 36(1), 332-341. https://doi.org/10.1609/icaps.v36i1.42843

Vancouver

Tuisov A, Vernik Y, Shleyfman A. Successor-Generator Planning with LLM-generated Heuristics. Proceedings of the International Conference on Automated Planning and Scheduling. 2026;36(1):332-341. doi:10.1609/icaps.v36i1.42843

BibTeX
@article{Tuisov2026,
  author = {Tuisov, Alexander and Vernik, Yonatan and Shleyfman, Alexander},
  title = {Successor-Generator Planning with LLM-generated Heuristics},
  journal = {Proceedings of the International Conference on Automated Planning and Scheduling},
  year = {2026},
  volume = {36},
  number = {1},
  pages = {332--341},
  doi = {10.1609/icaps.v36i1.42843}
}
Kostenlos starten →