Metraipolis › archive › dc_2da1bfe2a37e
Action- and Language-Conditioned Video Assessment for Embodied Control — arXiv:2608.08273
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.08273
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T09:12:32.262Z
- SHA-256 of the archived text
ca3c3993b525bf09765512412d395f1332821db6cf0b6ed221213a2370635a70- Stable URL of this page
https://metraipolis.com/documento/dc_2da1bfe2a37e
How to cite
Action- and Language-Conditioned Video Assessment for Embodied Control — arXiv:2608.08273 — arXiv (en), https://arxiv.org/abs/2608.08273 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2da1bfe2a37e (Metraipolis archive dc_2da1bfe2a37e, sha256 ca3c3993b525bf09765512412d395f1332821db6cf0b6ed221213a2370635a70)
BibTeX
@misc{dc2da1bfe2a37e,
title = {Action- and Language-Conditioned Video Assessment for Embodied Control — arXiv:2608.08273},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.08273},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2da1bfe2a37e; sha256 ca3c3993b525bf09765512412d395f1332821db6cf0b6ed221213a2370635a70},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.08273. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-based embodied agents executing multi-step natural language instructions require feedback mechanisms that assess task progress over complete trajectories. Conventional approaches based on final-frame matching or continuous embedding similarity may overlook intermediate transitions that are necessary for determining whether an instruction has been completed. We propose ALVA (Action- and Language-Conditioned Video Assessment), a trajectory evaluator that conditions its assessment on visual observations, the executed action sequence, and the natural language instruction. The method uses a pre-trained vision-language model (VLM) in two stages: it first summarizes frame-to-frame visual transitions conditioned on the executed actions and then assesses the generated summary with respect to the instruction to produce a discrete trajectory-level progress score. In simulated 3D household environments, ALVA exhibits a conservative assessment pattern with near-zero false-positive rates. When used as terminal feedback for closed-loop policy optimization, it provides more effective feedback than the evaluated static image and embedding-based visual baselines and reduces the performance gap to a ground-truth oracle. These results support action- and language-conditioned video assessment as an interpretable feedback mechanism for the evaluated simulated embodied-control tasks. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-11 in cs.RO). Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.08273 — 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 2026-10-19T09:11:58.741Z 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).