Metraipolis › archive › dc_81c1cc2cbdca
TacSushi: Tactile-Grounded World-Action Modeling for Dexterous Sushi Manipulation — arXiv:2609.19613
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.19613
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T03:23:26.213Z
- SHA-256 of the archived text
d6c1812e7f3773161769f2fba2b02b8a8e112117674a692235133e0442c94377- Stable URL of this page
https://metraipolis.com/documento/dc_81c1cc2cbdca
How to cite
TacSushi: Tactile-Grounded World-Action Modeling for Dexterous Sushi Manipulation — arXiv:2609.19613 — arXiv (en), https://arxiv.org/abs/2609.19613 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_81c1cc2cbdca (Metraipolis archive dc_81c1cc2cbdca, sha256 d6c1812e7f3773161769f2fba2b02b8a8e112117674a692235133e0442c94377)
BibTeX
@misc{dc81c1cc2cbdca,
title = {TacSushi: Tactile-Grounded World-Action Modeling for Dexterous Sushi Manipulation — arXiv:2609.19613},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.19613},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_81c1cc2cbdca; sha256 d6c1812e7f3773161769f2fba2b02b8a8e112117674a692235133e0442c94377},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19613. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Dexterous food manipulation requires control under deformation, occlusion, and uncertain contact. We present TacSushi, a tactile-grounded, Cosmos3-based world-action policy that learns from recorded future consequences while acting on current observations. The backbone encodes current RGB, language, and hand state, and feature-wise gated fusion incorporates fingertip tactile features into the action representation. During training, a decoder conditioned on demonstrated action chunks predicts logged future visual observations, task progress, relative contact risk, and tactile summaries; this decoder is removed at deployment. Failed trials provide consequence supervision, but their actions are excluded from imitation. We train TacSushi on 340 successful and 50 failed real-robot trials and compare six methods in 600 separate rollouts across three in-distribution tasks and two out-of-distribution ingredient variants. To assess food quality beyond a single geometric threshold, we score terminal outcomes using an anchored visual-quality protocol that equally weights five human ratings and three vision-language-model ratings per rollout. Full TacSushi achieves 68.3% average in-distribution success and 37.5% out-of-distribution success, compared with 36.7%/10.0% without future-consequence supervision and 25.0%/17.5% with direct tactile concatenation in place of gated fusion. These comparisons support complementary benefits of feature-wise gated tactile fusion and training-only predictive supervision. 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); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.19613 — 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 2026-10-10T03:04:13.840Z 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).