Metraipolis › archive › dc_71f886d7bd63

PREM: Prediction-Regulated Memory for Robot Manipulation — arXiv:2611.00634

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.00634
License (as archived)
CC0-1.0
Archived
2026-11-03T11:43:40.032Z
SHA-256 of the archived text
3779562af4004befcacbbd7e1c132b32a30f2e3120add7e7d8f436b12b0c617c
Stable URL of this page
https://metraipolis.com/documento/dc_71f886d7bd63

How to cite

PREM: Prediction-Regulated Memory for Robot Manipulation — arXiv:2611.00634 — arXiv (en), https://arxiv.org/abs/2611.00634 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_71f886d7bd63 (Metraipolis archive dc_71f886d7bd63, sha256 3779562af4004befcacbbd7e1c132b32a30f2e3120add7e7d8f436b12b0c617c)
BibTeX
@misc{dc71f886d7bd63,
  title = {PREM: Prediction-Regulated Memory for Robot Manipulation — arXiv:2611.00634},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00634},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_71f886d7bd63; sha256 3779562af4004befcacbbd7e1c132b32a30f2e3120add7e7d8f436b12b0c617c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00634. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A robot's next action can depend on recent motion or earlier visual evidence absent from its current observation. A recurrent visual-memory architecture can retain this history in a visual state, but its usefulness depends on what the state represents and how it is updated with new observations. We introduce Prediction-Regulated Memory (PREM), a framework that uses action-conditioned future prediction to guide both memory learning and updating in vision-language-action policies. During training, predicting future visual features provides temporal supervision for memory learning. During execution, PREM compares current visual features with the prediction made at the preceding decision and uses the spatially normalized prediction discrepancy to regulate updates to the visual state at each spatial location. The updated visual state supplies temporal context for action generation alongside current observations. Built on $\pi_{0.5}$, PREM improves average success rate from 9.63% to 20.74% on DOMINO dynamic manipulation tasks and from 14.4% to 83.8% on the RMBench M(1) memory-dependent setting. Further evaluations on LIBERO and LIBERO-Pro show improvements in general manipulation and robustness to spatial perturbations. PREM incurs a mean inference overhead of 7.7 ms on a single NVIDIA H800 GPU.

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/2611.00634 — 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 2026-11-03T11:06:37.675Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).