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

2026-09-14

AI Summary

针对自动驾驶中视觉-语言-动作模型推理与行为脱节问题，提出GRAVA框架，通过统一接地、推理和行动生成流程，增强物理场景证据与执行行为的关联。

Abstract

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.

Problem

• Grounded Reasoning
• Action Generation
• Visual-Language-Action Models
• Autonomous Driving
• Physical Evidence

Innovation

• Grounded Reasoning-to-Action
• autoregressive stream
• trajectory-anchored typed graph
• Executable Planner action
• agentic GRA data construction