Metraipolis › archive › dc_bca42860f8f4

Robota-Embed: Unified VLA Embeddings for Robot In-Context Learning and Evolution — arXiv:2610.19495

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19495
License (as archived)
CC0-1.0
Archived
2026-10-15T21:57:40.459Z
SHA-256 of the archived text
34598580e1ec1261acd18484fb1b3e02ce8e2fbecf133438e9b95234b33a84c3
Stable URL of this page
https://metraipolis.com/documento/dc_bca42860f8f4

How to cite

Robota-Embed: Unified VLA Embeddings for Robot In-Context Learning and Evolution — arXiv:2610.19495 — arXiv (en), https://arxiv.org/abs/2610.19495 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bca42860f8f4 (Metraipolis archive dc_bca42860f8f4, sha256 34598580e1ec1261acd18484fb1b3e02ce8e2fbecf133438e9b95234b33a84c3)
BibTeX
@misc{dcbca42860f8f4,
  title = {Robota-Embed: Unified VLA Embeddings for Robot In-Context Learning and Evolution — arXiv:2610.19495},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19495},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bca42860f8f4; sha256 34598580e1ec1261acd18484fb1b3e02ce8e2fbecf133438e9b95234b33a84c3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19495. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieving useful context for a robot requires matching its task, execution stage, and the continuation that should follow. We introduce Robota-Embed, to our knowledge the first vision-language-action (VLA) embedding model to jointly encode all three modalities in both current-state queries and recorded future continuations. Predictive retrieval training aligns these endpoints, enabling historical continuations to serve as context at inference time. We construct a multi-source corpus of 1,695,684 cleaned training anchors and introduce Robota-Embed-Bench, with 4,000 queries and a shared 8,000-entry gallery. Robota-Embed achieves 47.33% clip-level and 81.77% episode-level R@1, exceeding action-free Gemini Embedding 2 by 22.09 and 33.47 percentage points. Controlled retrieval studies show that removing native action degrades retrieval more than removing vision. We study two complementary applications. Embed2ICL uses retrieved experience for training-free and training-based policy conditioning, increasing success from 47.0% to 56.3% in DROID-Sim and from 86.4% to 92.8% on the LIBERO difficult subset, while improving task progress on real robots. Embed2Evolve uses execution feedback to grow task-local memory and refine context use around a frozen GPT-6 Astra policy. Across five ManiSkill tasks, it achieves 68% success, compared with 24% for GPT-6 Astra Direct and 33% for a harness independently evolved with Omni-Embed-Nemotron-3B under the same protocol. Relative to the latter, it uses approximately 79.5% fewer policy tokens per test rollout, counting both successes and failures. These results show that a unified, action-aware retrieval space can support robot in-context learning and task-local context evolution.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.19495 — 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 2026-10-15T21:38:29.565Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).