Self-Improvement for Fast, High-Quality Plan Generation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Gieselmann, Robert, von Huelsen, Henrike, Samson, Mihai, Meyer, Marie-Christine, Piotrowski, Dariusz, Radomskyi, Oleksandr, Okamoto, Justin, Gojayev, Turan, Painter, Michael, Brown, Gavin, Pecora, Federico, Wyatt, Jeremy L.
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910191970156544
author Gieselmann, Robert
von Huelsen, Henrike
Samson, Mihai
Meyer, Marie-Christine
Piotrowski, Dariusz
Radomskyi, Oleksandr
Okamoto, Justin
Gojayev, Turan
Painter, Michael
Brown, Gavin
Pecora, Federico
Wyatt, Jeremy L.
author_facet Gieselmann, Robert
von Huelsen, Henrike
Samson, Mihai
Meyer, Marie-Christine
Piotrowski, Dariusz
Radomskyi, Oleksandr
Okamoto, Justin
Gojayev, Turan
Painter, Michael
Brown, Gavin
Pecora, Federico
Wyatt, Jeremy L.
contents Generative models trained on synthetic plan data are a promising approach to generalized planning. Recent work has focused on finding any valid plan, rather than a high-quality solution. We address the challenge of producing high-quality plans, a computationally hard problem, in sub-exponential time. First, we demonstrate that, given optimal data, a decoder-only transformer can generate high-quality plans for unseen problem instances. Second, we show how to self-improve an initial model trained on sub-optimal data. Each round of self-improvement combines multiple model calls with graph search to generate improved plans, used for model fine-tuning. An experimental study on four domains: Blocksworld, Logistics, Labyrinth, and Sokoban, shows on average a 30% reduction in plan length over the source symbolic planner, with over 80% of plans being optimal, where the optimum is known. Plan quality is further improved by inference-time search. The model's latency scales sub-exponentially in contrast to the satisficing and optimal symbolic planners to which we compare. Together, these results suggest that self-improvement with generative models offers a scalable approach for high-quality plan generation.
format Preprint
id arxiv_https___arxiv_org_abs_2605_03625
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Self-Improvement for Fast, High-Quality Plan Generation
Gieselmann, Robert
von Huelsen, Henrike
Samson, Mihai
Meyer, Marie-Christine
Piotrowski, Dariusz
Radomskyi, Oleksandr
Okamoto, Justin
Gojayev, Turan
Painter, Michael
Brown, Gavin
Pecora, Federico
Wyatt, Jeremy L.
Artificial Intelligence
Generative models trained on synthetic plan data are a promising approach to generalized planning. Recent work has focused on finding any valid plan, rather than a high-quality solution. We address the challenge of producing high-quality plans, a computationally hard problem, in sub-exponential time. First, we demonstrate that, given optimal data, a decoder-only transformer can generate high-quality plans for unseen problem instances. Second, we show how to self-improve an initial model trained on sub-optimal data. Each round of self-improvement combines multiple model calls with graph search to generate improved plans, used for model fine-tuning. An experimental study on four domains: Blocksworld, Logistics, Labyrinth, and Sokoban, shows on average a 30% reduction in plan length over the source symbolic planner, with over 80% of plans being optimal, where the optimum is known. Plan quality is further improved by inference-time search. The model's latency scales sub-exponentially in contrast to the satisficing and optimal symbolic planners to which we compare. Together, these results suggest that self-improvement with generative models offers a scalable approach for high-quality plan generation.
title Self-Improvement for Fast, High-Quality Plan Generation
topic Artificial Intelligence
url https://arxiv.org/abs/2605.03625