Atomic Action Slicing: Planner-Aligned Options for Generalist VLA Agents
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918246536445952 |
|---|---|
| author | Tabakov, Stefan Popov, Asen Dimitrov, Dimitar Kiyamousavi, S. Ensiye Hristov, Vladimir Kraychev, Boris |
| author_facet | Tabakov, Stefan Popov, Asen Dimitrov, Dimitar Kiyamousavi, S. Ensiye Hristov, Vladimir Kraychev, Boris |
| contents | 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) |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_11584 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Atomic Action Slicing: Planner-Aligned Options for Generalist VLA Agents Tabakov, Stefan Popov, Asen Dimitrov, Dimitar Kiyamousavi, S. Ensiye Hristov, Vladimir Kraychev, Boris Machine Learning Artificial Intelligence Robotics 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) |
| title | Atomic Action Slicing: Planner-Aligned Options for Generalist VLA Agents |
| topic | Machine Learning Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2512.11584 |