SkillVLA: Tackling Combinatorial Diversity in Dual-Arm Manipulation via Skill Reuse

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhai, Xuanran, Huang, Zekai, Wu, Longyan, Zhao, Qianyou, Yu, Qiaojun, Ren, Jieji, Hao, Ce, Soh, Harold
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912942156414976
author Zhai, Xuanran
Huang, Zekai
Wu, Longyan
Zhao, Qianyou
Yu, Qiaojun
Ren, Jieji
Hao, Ce
Soh, Harold
author_facet Zhai, Xuanran
Huang, Zekai
Wu, Longyan
Zhao, Qianyou
Yu, Qiaojun
Ren, Jieji
Hao, Ce
Soh, Harold
contents Recent progress in vision-language-action (VLA) models has demonstrated strong potential for dual-arm manipulation, enabling complex behaviors and generalization to unseen environments. However, mainstream bimanual VLA formulations largely overlook the critical challenge of combinatorial diversity. Different pairings of single-arm behaviors can induce qualitatively distinct task behaviors, yet existing models do not explicitly account for this structure. We argue that effective bimanual VLAs should support skill reuse - the ability to recombine previously learned single-arm skills across novel left-right pairings - thereby avoiding the need to separately learn every possible combination. Current VLA designs entangle skills across arms, preventing such recomposition and limiting scalability. To address this limitation, we propose SkillVLA, a framework explicitly designed to enable skill reuse in dual-arm manipulation. Extensive experiments demonstrate that SkillVLA substantially improves skill composition, increasing overall success rate from 0% to 51%, and achieves strong performance on cooperative and long-horizon tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03836
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SkillVLA: Tackling Combinatorial Diversity in Dual-Arm Manipulation via Skill Reuse
Zhai, Xuanran
Huang, Zekai
Wu, Longyan
Zhao, Qianyou
Yu, Qiaojun
Ren, Jieji
Hao, Ce
Soh, Harold
Robotics
Recent progress in vision-language-action (VLA) models has demonstrated strong potential for dual-arm manipulation, enabling complex behaviors and generalization to unseen environments. However, mainstream bimanual VLA formulations largely overlook the critical challenge of combinatorial diversity. Different pairings of single-arm behaviors can induce qualitatively distinct task behaviors, yet existing models do not explicitly account for this structure. We argue that effective bimanual VLAs should support skill reuse - the ability to recombine previously learned single-arm skills across novel left-right pairings - thereby avoiding the need to separately learn every possible combination. Current VLA designs entangle skills across arms, preventing such recomposition and limiting scalability. To address this limitation, we propose SkillVLA, a framework explicitly designed to enable skill reuse in dual-arm manipulation. Extensive experiments demonstrate that SkillVLA substantially improves skill composition, increasing overall success rate from 0% to 51%, and achieves strong performance on cooperative and long-horizon tasks.
title SkillVLA: Tackling Combinatorial Diversity in Dual-Arm Manipulation via Skill Reuse
topic Robotics
url https://arxiv.org/abs/2603.03836