ChatHTN: Interleaving Approximate (LLM) and Symbolic HTN Planning

Héctor Muñoz-Avila, David W. Aha, Paola Rizzo
Proceedings of the International Conference on Neuro-symbolic Systems, PMLR 288:446-458, 2025.

Abstract

We introduce ChatHTN, a Hierarchical Task Network (HTN) planner that combines symbolic HTN planning techniques with queries to ChatGPT to approximate solutions in the form of task decompositions. The resulting hierarchies interleave task decompositions generated by symbolic HTN planning with those generated by ChatGPT. Despite the approximate nature of the results generated by ChatGPT, ChatHTN is provably sound; any plan it generates correctly achieves the input tasks. We demonstrate this property with an open-source implementation of our system.

Cite this Paper

BibTeX
@InProceedings{pmlr-v288-munoz-avila25a, title = {ChatHTN: Interleaving Approximate (LLM) and Symbolic HTN Planning}, author = {Mu\~noz-Avila, H\'ector and Aha, David W. and Rizzo, Paola}, booktitle = {Proceedings of the International Conference on Neuro-symbolic Systems}, pages = {446--458}, year = {2025}, editor = {Pappas, George and Ravikumar, Pradeep and Seshia, Sanjit A.}, volume = {288}, series = {Proceedings of Machine Learning Research}, month = {28--30 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v288/main/assets/munoz-avila25a/munoz-avila25a.pdf}, url = {https://proceedings.mlr.press/v288/munoz-avila25a.html}, abstract = {We introduce ChatHTN, a Hierarchical Task Network (HTN) planner that combines symbolic HTN planning techniques with queries to ChatGPT to approximate solutions in the form of task decompositions. The resulting hierarchies interleave task decompositions generated by symbolic HTN planning with those generated by ChatGPT. Despite the approximate nature of the results generated by ChatGPT, ChatHTN is provably sound; any plan it generates correctly achieves the input tasks. We demonstrate this property with an open-source implementation of our system.} }
Endnote
%0 Conference Paper %T ChatHTN: Interleaving Approximate (LLM) and Symbolic HTN Planning %A Héctor Muñoz-Avila %A David W. Aha %A Paola Rizzo %B Proceedings of the International Conference on Neuro-symbolic Systems %C Proceedings of Machine Learning Research %D 2025 %E George Pappas %E Pradeep Ravikumar %E Sanjit A. Seshia %F pmlr-v288-munoz-avila25a %I PMLR %P 446--458 %U https://proceedings.mlr.press/v288/munoz-avila25a.html %V 288 %X We introduce ChatHTN, a Hierarchical Task Network (HTN) planner that combines symbolic HTN planning techniques with queries to ChatGPT to approximate solutions in the form of task decompositions. The resulting hierarchies interleave task decompositions generated by symbolic HTN planning with those generated by ChatGPT. Despite the approximate nature of the results generated by ChatGPT, ChatHTN is provably sound; any plan it generates correctly achieves the input tasks. We demonstrate this property with an open-source implementation of our system.
APA
Muñoz-Avila, H., Aha, D.W. & Rizzo, P.. (2025). ChatHTN: Interleaving Approximate (LLM) and Symbolic HTN Planning. Proceedings of the International Conference on Neuro-symbolic Systems, in Proceedings of Machine Learning Research 288:446-458 Available from https://proceedings.mlr.press/v288/munoz-avila25a.html.

Related Material