Metraipolis › archive › dc_3e75e8343cb3

Cue the Flow: Steering Flow-Matching Policies for Open-World Delivery Manipulation — arXiv:2609.38989

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.38989
License (as archived)
CC0-1.0
Archived
2026-10-07T15:11:48.868Z
SHA-256 of the archived text
3829d5f37ceacc1752514a4c8dd71132e5ea22933c7094f536ffcb3e5131ec35
Stable URL of this page
https://metraipolis.com/documento/dc_3e75e8343cb3

How to cite

Cue the Flow: Steering Flow-Matching Policies for Open-World Delivery Manipulation — arXiv:2609.38989 — arXiv (en), https://arxiv.org/abs/2609.38989 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3e75e8343cb3 (Metraipolis archive dc_3e75e8343cb3, sha256 3829d5f37ceacc1752514a4c8dd71132e5ea22933c7094f536ffcb3e5131ec35)
BibTeX
@misc{dc3e75e8343cb3,
  title = {Cue the Flow: Steering Flow-Matching Policies for Open-World Delivery Manipulation — arXiv:2609.38989},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38989},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3e75e8343cb3; sha256 3829d5f37ceacc1752514a4c8dd71132e5ea22933c7094f536ffcb3e5131ec35},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38989. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Open-world goods delivery requires mobile manipulators to follow free-form user instructions and manipulate potentially novel objects. Existing dual-system approaches use high-level grounding models to convert language into grounded visual prompts, but their low-level controllers can remain brittle under noisy perception, dynamic scenes, and contact-rich interactions. We instead use a pretrained flow-matching vision-language-action model as the low-level control interface, leveraging its reactivity and robustness to environmental changes while treating the grounding output as a spatial cue for policy steering. Our key insight is that the pretrained VLA already provides a strong manipulation prior, while the spatial cue supplies the missing target information needed to guide actions under novel language--object mappings. Concretely, we introduce a lightweight cue-conditioned adapter. The adapter is first trained with contrastive objectives to produce salient and spatially discriminative cue representations, and is then supervised to predict a diagonal affine transformation over the generated action chunk, aligning policy steering with the cued target. Across tabletop and mobile-base settings, our method improves instruction following and manipulation success on both in-domain and out-of-domain objects, achieving up to near $2\times$ improvement in average task success rate with negligible inference overhead.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.38989 — 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 2026-10-07T14:44:47.398Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).