Metraipolis › archive › dc_ad4334fd98cb

VABench: Measuring Embodied Spatial Intelligence through Visual Demonstrations, Active Perception, and Metric Control — arXiv:2609.19554

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.19554
License (as archived)
CC0-1.0
Archived
2026-10-10T05:08:08.603Z
SHA-256 of the archived text
dfc2bca59087cfbe73aa8ca40080c4b78f8aba66bdd9eac24f441846ee33fad9
Stable URL of this page
https://metraipolis.com/documento/dc_ad4334fd98cb

How to cite

VABench: Measuring Embodied Spatial Intelligence through Visual Demonstrations, Active Perception, and Metric Control — arXiv:2609.19554 — arXiv (en), https://arxiv.org/abs/2609.19554 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ad4334fd98cb (Metraipolis archive dc_ad4334fd98cb, sha256 dfc2bca59087cfbe73aa8ca40080c4b78f8aba66bdd9eac24f441846ee33fad9)
BibTeX
@misc{dcad4334fd98cb,
  title = {VABench: Measuring Embodied Spatial Intelligence through Visual Demonstrations, Active Perception, and Metric Control — arXiv:2609.19554},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19554},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ad4334fd98cb; sha256 dfc2bca59087cfbe73aa8ca40080c4b78f8aba66bdd9eac24f441846ee33fad9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19554. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Spatial intelligence requires more than describing object locations. Under incomplete observation, models must identify and acquire missing evidence, interpret it in a common spatial frame, and act on it. We introduce VA-Bench to evaluate the complete observe-reason-act-revise loop. General-purpose MLLMs learn procedural context from RGB-only demonstrations, actively select camera viewpoints, issue metric Cartesian commands, and revise them from execution feedback. Models receive no privileged object poses, oracle trajectories, or learned action heads. A fixed model-agnostic controller executes only model-specified targets. VA-Bench contains 14 base task families (11 single-arm and three dual-arm), seven held-out geometry/layout variants, and a long-horizon five-object composition track. We evaluate 12 primary model conditions in three independent runs over the same 20 physically verified seeds per base task, reporting terminal success, nine trajectory-level behavioral diagnostics, and subtask progress. First, the best-performing model scores 100.0% on target localization and 78.9% on spatial relations in the annotated run. Its three-run macro-average task success is only 53.93+/-3.17%. Second, active camera control significantly improves task success over passive multi-view observation. In one matched comparison, success rises from 27.86% to 57.50%. Third, held-out geometric transfer can reduce task success by over 30 percentage points. No model completes a strict long-horizon episode, despite substantial partial progress. VA-Bench thus tests whether general-purpose MLLMs can turn visual demonstrations and actively acquired evidence into successful embodied action.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-18 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.19554 — 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 2026-10-10T04:26:09.639Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).