Geo-VLA: Geometry-Aware Vision-Language-Action Planning via Internalization of Map Semantics
TL;DR
Geo-VLA is proposed, a plug-and-play framework that enhances VLA models by learning geometry-aware visual representations and introduces Geo-QA, a geometry-focused question-answering dataset that injects road geometry into vision-language representations through contrastive learning and instruction tuning.
Abstract
Vision-language-action (VLA) models have advanced end-to-end autonomous driving by leveraging foundation models for semantic reasoning and long-tail generalization. However, their planning performance remains limited in complex driving environments because image-only representations inadequately capture planning-relevant road geometry and topology. In this paper, we propose Geo-VLA, a plug-and-play framework that enhances VLA models by learning geometry-aware visual representations. During training, Geo-VLA internalizes geometric map semantics to strengthen road-structure representations, while requiring no HD maps or additional lane information during inference. To support this approach, we introduce Geo-QA, a geometry-focused question-answering dataset that injects road geometry into vision-language representations through contrastive learning and instruction tuning. Experiments on NAVSIM v1 demonstrate that Geo-VLA consistently improves VLA planners with distinct action-generation architectures, achieving 92.1 PDMS and establishing a new state-of-the-art among single-camera VLA planners.