Metraipolis › archive › dc_22ce0beb8b79

Skills in Weights, Memory in Code: Hybrid Learning for Memory-Dependent Robot Manipulation — arXiv:2608.09410

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.09410
License (as archived)
CC0-1.0
Archived
2026-10-19T09:12:47.785Z
SHA-256 of the archived text
79b3227a559df2ecec17019e090d6ce817a97155df8bafb720c6c648d034ea9a
Stable URL of this page
https://metraipolis.com/documento/dc_22ce0beb8b79

How to cite

Skills in Weights, Memory in Code: Hybrid Learning for Memory-Dependent Robot Manipulation — arXiv:2608.09410 — arXiv (en), https://arxiv.org/abs/2608.09410 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_22ce0beb8b79 (Metraipolis archive dc_22ce0beb8b79, sha256 79b3227a559df2ecec17019e090d6ce817a97155df8bafb720c6c648d034ea9a)
BibTeX
@misc{dc22ce0beb8b79,
  title = {Skills in Weights, Memory in Code: Hybrid Learning for Memory-Dependent Robot Manipulation — arXiv:2608.09410},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.09410},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_22ce0beb8b79; sha256 79b3227a559df2ecec17019e090d6ce817a97155df8bafb720c6c648d034ea9a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.09410. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern vision-language-action (VLA) policies have acquired broad manipulation skills, but typically generate each action chunk from the current observation or a short fixed-length history. However, real-world manipulation is often non-Markovian, requiring robots to retain and reason over task-relevant information from long-horizon interaction histories to determine the next action. To address this challenge, we propose HyMeS, a hybrid learning framework that leverages the reasoning and memory-management capabilities of coding agents to steer a Markovian VLA for memory-dependent manipulation. Specifically, HyMeS learns low-level motor skills through gradient-based imitation learning, while a coding agent acquires high-level memory-management strategies through heuristic learning by iteratively updating an executable heuristic system from rollout feedback. Furthermore, we close the loop between steering and execution through multimodal stage-completion verification, which updates memory using proprioceptive signals and multi-frame VLM judgments. Compared with end-to-end memory-augmented VLAs, HyMeS requires demonstrations only for reusable motor skills rather than for every history-dependent task configuration, enabling data-efficient compositional generalization. On RoboMemArena, HyMeS improves mean cumulative success from 52.5% to 66.2% and mean task success from 41.3% to 60.1% over pi0.5, while outperforming PrediMem by 4.5 points in cumulative success and 14.5 points in task success.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-11 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2608.09410 — 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 2026-10-19T09:11:58.783Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).