Current vision-language-action (VLA) models generalize poorly, particularly when tasks require new compositions of skills or objects. We introduce Atomic Action Slicing (AAS), a planner-aligned approach that decomposes long-horizon demonstrations into short, typed atomic actions that are easier for planners to use and policies to learn. Using LIBERO demonstrations, AAS produces a validated dataset of 2,124 atomic segments labeled with action type, temporal span, and confidence. A stronger segmenter (Gemini 2.5 Pro) closely matches planner-defined plans and remains robust under keyframe jitter, while smaller models perform worse on multi-object tasks. Fine-tuning CLIP-RT+ on our atomic dataset improves task success from 94.2% to 95.3% on LIBERO-Goal and 83.8% to 88.8% on LIBERO-Long. We publicly release the GATE-VLAP dataset on HuggingFace(https://huggingface.co/datasets/gate-institute/GATE-VLAP-datasets)
@article{arxiv.2512.11584,
title = {Atomic Action Slicing: Planner-Aligned Options for Generalist VLA Agents},
author = {Stefan Tabakov and Asen Popov and Dimitar Dimitrov and S. Ensiye Kiyamousavi and Vladimir Hristov and Boris Kraychev},
journal= {arXiv preprint arXiv:2512.11584},
year = {2025}
}
Comments
The 41st ACM/SIGAPP Symposium On Applied Computing