Metraipolis › archive › dc_a1ed375ce4e0
From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models — arXiv:2610.39543
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.39543
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-01T02:44:27.419Z
- SHA-256 of the archived text
d4bd43b85d9b9e8d99e486fce384a67b9d0d709b87dc03343ba8f3f28ff3aa41- Stable URL of this page
https://metraipolis.com/documento/dc_a1ed375ce4e0
How to cite
From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models — arXiv:2610.39543 — arXiv (en), https://arxiv.org/abs/2610.39543 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a1ed375ce4e0 (Metraipolis archive dc_a1ed375ce4e0, sha256 d4bd43b85d9b9e8d99e486fce384a67b9d0d709b87dc03343ba8f3f28ff3aa41)
BibTeX
@misc{dca1ed375ce4e0,
title = {From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models — arXiv:2610.39543},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.39543},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a1ed375ce4e0; sha256 d4bd43b85d9b9e8d99e486fce384a67b9d0d709b87dc03343ba8f3f28ff3aa41},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39543. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reasoning informs decision-making by connecting intended actions with their expected consequences. Recent vision-language-action (VLA) models incorporate embodied chains of thought (ECoTs) to guide action generation. However, such guidance remains implicit, as actions are conditioned on ECoTs rather than explicitly optimized to align with them. We introduce CoRAct, a consequence-grounded reasoning-to-action alignment framework that explicitly aligns reasoning and action through reward-guided optimization in action noise space. CoRAct formulates two physically grounded ECoT components as verifiable signals: movement specifies intended motion, while waypoints encode expected spatial consequences. A unified Mixture-of-Transformers jointly models ECoT reasoning, action generation, and action-conditioned future prediction. For candidate actions generated from sampled noise, CoRAct computes alignment rewards based on movement consistency and agreement between predicted outcomes and specified waypoints. These rewards guide group-relative optimization of the initial noise distribution, steering action generation toward consistency with intended actions and expected consequences. Experiments across RoboCasa, VLABench, and real-world manipulation tasks demonstrate improvements in task success, generalization, and reasoning-execution alignment over strong ECoT-based VLA baselines. CoRAct also grounds externally provided ECoTs in executable behavior, enabling zero-shot adaptation to novel scenarios by leveraging knowledge from powerful vision-language models. Project page and dataset: this https URL Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2610.39543 — 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 2026-11-01T02:16:58.662Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).