Metraipolis › archive › dc_c755dbb09060
Scaling Bimanual Household Manipulation from 1,500 hours of Demonstrations to On-Policy Corrections — arXiv:2609.03591
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.03591
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T02:57:28.726Z
- SHA-256 of the archived text
c02640ec3828114720b726c14bc07e8845fa6e8bd903050b7fea77ad8ac92ee0- Stable URL of this page
https://metraipolis.com/documento/dc_c755dbb09060
How to cite
Scaling Bimanual Household Manipulation from 1,500 hours of Demonstrations to On-Policy Corrections — arXiv:2609.03591 — arXiv (en), https://arxiv.org/abs/2609.03591 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c755dbb09060 (Metraipolis archive dc_c755dbb09060, sha256 c02640ec3828114720b726c14bc07e8845fa6e8bd903050b7fea77ad8ac92ee0)
BibTeX
@misc{dcc755dbb09060,
title = {Scaling Bimanual Household Manipulation from 1,500 hours of Demonstrations to On-Policy Corrections — arXiv:2609.03591},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.03591},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c755dbb09060; sha256 c02640ec3828114720b726c14bc07e8845fa6e8bd903050b7fea77ad8ac92ee0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03591. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Learning generalist policies for robust bimanual manipulation is bottlenecked by the scarcity of high quality large scale human demonstration data. In this work, we release 1,500 hours of diverse bimanual manipulation demonstrations covering everyday household tasks, and use this comprehensive corpus to train XR-2, a powerful vision-language-action (VLA) model. Enabled by a purpose built high throughput data pipeline and a carefully designed multi stage training paradigm, XR-2 attains strong manipulation performance in our systematic experiments while retaining favorable training efficiency and high data utilization. We further study two critical scaling axes: varying the amount of expert demonstration data, and post training on DAgger correction data from real time human interventions. In both settings, task success rate improves steadily over the data ranges we probe, exhibiting a clear consistent scaling trend at our current data scale. These results validate both the learning capacity of XR-2 and the promising scaling properties of the released dataset, which we open source to support reproducible research on bimanual robot manipulation learning. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-04 in cs.RO). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.03591 — 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 2026-10-17T02:33:25.387Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).