Metraipolis › archive › dc_ded4287022f2

Copper-Policy: Focus on the Representation for Robust Robot Manipulation — arXiv:2609.32779

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32779
License (as archived)
CC0-1.0
Archived
2026-10-07T22:39:46.641Z
SHA-256 of the archived text
622f2802934227a880c3aee6a0bfee6f9245efd5b6a925d27d3eeaed62477ac8
Stable URL of this page
https://metraipolis.com/documento/dc_ded4287022f2

How to cite

Copper-Policy: Focus on the Representation for Robust Robot Manipulation — arXiv:2609.32779 — arXiv (en), https://arxiv.org/abs/2609.32779 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ded4287022f2 (Metraipolis archive dc_ded4287022f2, sha256 622f2802934227a880c3aee6a0bfee6f9245efd5b6a925d27d3eeaed62477ac8)
BibTeX
@misc{dcded4287022f2,
  title = {Copper-Policy: Focus on the Representation for Robust Robot Manipulation — arXiv:2609.32779},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32779},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ded4287022f2; sha256 622f2802934227a880c3aee6a0bfee6f9245efd5b6a925d27d3eeaed62477ac8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32779. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

World Action Models (WAMs) acquire behavioral priors by modeling future scene evolution, but predicting detailed futures in pixel or latent space incurs substantial cost. Recent evidence that co-training gains persist without test-time generation raises a question: what must a WAM learn to improve control? We introduce Copper-Policy, which learns a compact World representation with the policy rather than relying on a predefined target space. Through temporal joint-embedding prediction, it predicts future observation embeddings conditioned on task intention without reconstructing pixels. This prediction and action decoding shape the representation jointly, while the policy retains access to current-frame spatial detail for execution. Representation analyses show that the learned features better separate task-driven change from perturbations and provide complementary information for control. Compact prediction targets reduce training tokens per sample, enabling a 2B-parameter model trained in 9.67 hours on 8$\times$ RTX 5090 GPUs and 6$\times$ faster than Fast-WAM on matched A100 GPUs. Copper-Policy outperforms every compared method without embodied pretraining on RoboTwin and several embodied-pretrained VLAs on LIBERO-Plus (80.85%). On three challenging real-robot tasks, it performs comparably to $\pi_{0.5}$ and attains a higher average score. Together, these results show that Copper-Policy combines strong control performance with efficient training.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.32779 — 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 2026-10-07T22:14:28.641Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).