Metraipolis › archive › dc_69038c9dfc8b

GRAVA: Grounded Reasoning-to-Action Representation and Learning for Autonomous Driving — arXiv:2609.15169

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.15169
License (as archived)
CC0-1.0
Archived
2026-10-16T01:53:23.644Z
SHA-256 of the archived text
c3a7b2ebf00ad0df278b5390f335bba5ee3ef4cb8c7bf93940e7f0248066ab3d
Stable URL of this page
https://metraipolis.com/documento/dc_69038c9dfc8b

How to cite

GRAVA: Grounded Reasoning-to-Action Representation and Learning for Autonomous Driving — arXiv:2609.15169 — arXiv (en), https://arxiv.org/abs/2609.15169 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_69038c9dfc8b (Metraipolis archive dc_69038c9dfc8b, sha256 c3a7b2ebf00ad0df278b5390f335bba5ee3ef4cb8c7bf93940e7f0248066ab3d)
BibTeX
@misc{dc69038c9dfc8b,
  title = {GRAVA: Grounded Reasoning-to-Action Representation and Learning for Autonomous Driving — arXiv:2609.15169},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.15169},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_69038c9dfc8b; sha256 c3a7b2ebf00ad0df278b5390f335bba5ee3ef4cb8c7bf93940e7f0248066ab3d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15169. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Driving vision-language-action (VLA) models increasingly reason before acting, but their intermediate reasoning is often weakly grounded in physical scene evidence and loosely connected to executable behavior. We present GRAVA, a framework built around Grounded Reasoning-to-Action (GRA), which unifies grounding, reasoning, and action generation in a single autoregressive stream. GRA links action-relevant language references to 2D visual regions and ego-centric physical states, organizes object interactions and decisions in a trajectory-anchored typed graph, and serializes this structure into grounded reasoning. A single VLM generates this reasoning followed by a compact Executable Planner action that is deterministically decoded into a continuous trajectory. We further introduce an agentic GRA data construction pipeline that combines forward scene grounding with backward trajectory anchoring, and use it to build GR-NavSim with 2.2M grounded question-answer pairs and 70K GRA reasoning traces. A progressive training strategy develops grounded cognition through pre-training, establishes the reasoning-to-action interface through imitation, and improves driving behavior through reinforcement learning and exploration. Using about 60% of the available human driving demonstrations for action supervision, GRAVA-8B achieves state-of-the-art performance among purely autoregressive driving models on the full NAVSIM benchmark. On an internal long-tail benchmark, full GRA improves key-object compliance and Closed-loop Driving Score by 19.3% and 20.5% over action-only prediction, respectively. These results show the benefit of preserving action-relevant physical evidence from grounded reasoning through executable action generation.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.15169 — 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 2026-10-16T01:25:23.646Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).