Metraipolis › archive › dc_3d400cb6318b
VLRLA: Deployment-Time Adaptation of Vision-Language-Action Models through In-Context Reinforcement Learning — arXiv:2610.19517
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.19517
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T20:01:42.678Z
- SHA-256 of the archived text
5931fb54db0f38904304af800da506193c6ce4b50c32b24505e4e276501f0bad- Stable URL of this page
https://metraipolis.com/documento/dc_3d400cb6318b
How to cite
VLRLA: Deployment-Time Adaptation of Vision-Language-Action Models through In-Context Reinforcement Learning — arXiv:2610.19517 — arXiv (en), https://arxiv.org/abs/2610.19517 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3d400cb6318b (Metraipolis archive dc_3d400cb6318b, sha256 5931fb54db0f38904304af800da506193c6ce4b50c32b24505e4e276501f0bad)
BibTeX
@misc{dc3d400cb6318b,
title = {VLRLA: Deployment-Time Adaptation of Vision-Language-Action Models through In-Context Reinforcement Learning — arXiv:2610.19517},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.19517},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3d400cb6318b; sha256 5931fb54db0f38904304af800da506193c6ce4b50c32b24505e4e276501f0bad},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19517. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language-action (VLA) models connect visual perception and language understanding with robot control, providing a foundation for instruction-conditioned manipulation. However, conventional pretrained VLA policies lack a mechanism for gradual adaptation through accumulated interaction experience during deployment. We therefore propose VLRLA, a general framework that equips existing VLA models with \textit{in-context reinforcement learning} capabilities for deployment-time adaptation. Our approach augments a VLA model with a reinforcement learning module that learns from historical observations, actions, rewards, and termination signals. During deployment, the policy gradually refines its action generation and progressively improves task success while keeping model parameters fixed. Our approach is also designed for general applicability and compatibility with diverse learning history construction strategies and model training pipelines. Finally, through extensive experiments, we systematically investigate the learning properties of VLRLA in simulation and demonstrate its ability to adapt to unseen scenarios during real-world deployment.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.19517 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T19:35:48.241Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).