Metraipolis › archive › dc_376bd576a24b
Improving Cross-embodiment Transfer in Latent Action Models with Action-Similarity Supervision — arXiv:2609.19846
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.19846
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T03:23:42.278Z
- SHA-256 of the archived text
f066ad75e2e5267b580a02e027ef693e8d128e7adfcc5738770fd82dd2367abe- Stable URL of this page
https://metraipolis.com/documento/dc_376bd576a24b
How to cite
Improving Cross-embodiment Transfer in Latent Action Models with Action-Similarity Supervision — arXiv:2609.19846 — arXiv (en), https://arxiv.org/abs/2609.19846 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_376bd576a24b (Metraipolis archive dc_376bd576a24b, sha256 f066ad75e2e5267b580a02e027ef693e8d128e7adfcc5738770fd82dd2367abe)
BibTeX
@misc{dc376bd576a24b,
title = {Improving Cross-embodiment Transfer in Latent Action Models with Action-Similarity Supervision — arXiv:2609.19846},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.19846},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_376bd576a24b; sha256 f066ad75e2e5267b580a02e027ef693e8d128e7adfcc5738770fd82dd2367abe},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19846. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
As generalist robot policies gain vision and language from web-scale pretraining, demonstrations remain costly to collect and tied to the robot that recorded them. Latent action models (LAMs) address both by learning latent actions from action-free videos that can be shared across embodiments, however, in practice, LAMs are sensitive to background visual noise, and the same motion from two different robots may be encoded with different latents. One solution to the background visual noise is to add an auxiliary loss predicting the robot action from the latent action, further associating the latent action space to the embodiment specific robot action space. We study a different use of the same labels, through action-similarity supervision. The similarity between any two latent actions is trained to match the similarity of the two ground-truth robot action sequences. The ground-truth actions are never predicted by the LAM, so the latent action does not need to encode embodiment specifics. We evaluate cross-embodiment transfer on RoboTwin 2.0 in a controlled setup, two bimanual robots demonstrate disjoint task sets, a policy is trained on all the demonstrations, and each robot is evaluated closed-loop on the tasks only the other demonstrated. With the policy architecture and its hyperparameters, the dataset, and the evaluation protocol fixed, predicting latent actions instead of ground-truth actions more than doubles cross-embodiment success. Given the same ground-truth actions, similarity supervision transfers better than an auxiliary loss that predicts the ground-truth action during the LAM training. Computing the similarities on end-effector motion rather than joint-space motion, and letting the loss compare latent actions across the two robots, gives the best approach of the study. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AI). Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.19846 — 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 2026-10-10T03:04:13.860Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).