This dataset was developed by a team from Skoltech's Intelligent Space Robotics Laboratory.
The dataset was used to train LLM for Robot Behavior Tree Generation based on a user command.
Note that this model is part of a multi-agent artificial intelligence system for the dog robot described in the LLM-MARS paper.
The dataset is split into parts by different game strategies
Paper preprint BibTeX cite:
@misc{lykov2023llmmarslargelanguagemodel,
title={LLM-MARS: Large Language Model for… See the full description on the dataset page:
https://huggingface.co/datasets/ArtemLykov/LLM-MARS_dataset.
