shadyy/ROBOCUP_24

ROBOCUP24@Home ##Abstract Large Language Models (LLMs) have been widely utilized to perform complex robotic tasks. However, handling task-specific complex planning based on robot knowledge and abilities remains a significant challenge. This paper proposes a novel method for dataset generation that integrates hand-annotated data with automated scripts to create a diverse, task-specific training dataset for fine-tuning LLMs. This approach not only enhances the diversity and volume… See the full description on the dataset page: https://huggingface.co/datasets/shadyy/ROBOCUP_24.