Metraipolis › archive › dc_cf53db5ed6a2
Active Generative Simulation: Meta-Learning What to Simulate for Sim-and-Real Policy Co-Training — arXiv:2610.19479
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.19479
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T21:57:50.241Z
- SHA-256 of the archived text
7baff45080adb6fd0c5f16cfd09cc778dbb317b6e4bc27f9d21af826639e64f5- Stable URL of this page
https://metraipolis.com/documento/dc_cf53db5ed6a2
How to cite
Active Generative Simulation: Meta-Learning What to Simulate for Sim-and-Real Policy Co-Training — arXiv:2610.19479 — arXiv (en), https://arxiv.org/abs/2610.19479 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cf53db5ed6a2 (Metraipolis archive dc_cf53db5ed6a2, sha256 7baff45080adb6fd0c5f16cfd09cc778dbb317b6e4bc27f9d21af826639e64f5)
BibTeX
@misc{dccf53db5ed6a2,
title = {Active Generative Simulation: Meta-Learning What to Simulate for Sim-and-Real Policy Co-Training — arXiv:2610.19479},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.19479},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cf53db5ed6a2; sha256 7baff45080adb6fd0c5f16cfd09cc778dbb317b6e4bc27f9d21af826639e64f5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19479. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Sim-and-real co-training is an effective way to scale robot learning by augmenting limited real-world demonstrations with simulation data. However, existing approaches typically treat simulators as fixed data sources, relying on uniform generation, offline retrieval, or manual parameter alignment by domain experts. We formulate \textit{Active Generative Simulation}: optimizing a simulator's data-generation distribution to maximize downstream real-world performance. To solve this problem, we propose MetaGenSim, a meta-learning framework that trains an auxiliary rater network via meta-gradients to automatically discover high-utility data-generation distributions within a user-specified simulation parameter space, without manual parameter alignment. Across simulation and real-world manipulation tasks, MetaGenSim outperforms uniform generation and retrieval baselines, achieves a 150% relative improvement in real-world full-task success over uniform generation, and matches expert manual alignment. We further show that distributions curated with a lightweight proxy policy transfer to VLA fine-tuning, and that MetaGenSim can exploit partial alignment through atomic skill stitching and object-receptacle composition when exact sim-and-real task alignment is unavailable. Project website: this https URL
Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.19479 — 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 2026-10-15T21:38:44.448Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).