DriveStack-VLA: Render-Teacher Alignment for BEV-Based DeepStack Vision-Language-Action Model

• URL: http://arxiv.org/abs/2606.24051v1
• Date: Tue, 23 Jun 2026 01:40:54 GMT
• Title: DriveStack-VLA: Render-Teacher Alignment for BEV-Based DeepStack Vision-Language-Action Model
• Abstract summary: We present DriveStack-VLA, a framework built upon a large VLM backbone.<n>We develop dual visual modeling components to strengthen the spatial grounding of VLA driving.<n>DriveStack-VLA achieves 91.6 PDMS on NAVSIMv1, 91.0S on NAVSIMv2, and a driving score of 79.49 with a success rate of 56.36% on the closed-loop Bench2Drive.
• Score: 22.06522773748522
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language-Action driving models convert a pretrained Vision-Language Model into a driving policy, allowing them to use world knowledge and follow language guidances. However, existing VLA driving models still lack driving-oriented spatial intelligence: their policies are mainly grounded on perspective image tokens and language priors, while precise motion planning requires metric geometry, top-down scene structure, and attention to safety-critical perceptual cues. This limitation makes current models vulnerable to weak visual geometry modeling and perceptual coverage in expert demonstrations. In this paper, we present DriveStack-VLA, a framework built upon a large VLM backbone. To strengthen the spatial grounding of VLA driving, we develop dual visual modeling components. We inject a Bird-Eye-View representation into the Large Language Model decoder through a DeepStack-style connection, and propose Render-Teacher Alignment to align the perceptual focus of real images with that of rasterized images. Furthermore, to bridge the gap in multimodal trajectory selection, we introduce a head-based self-critique module that ranks sampled trajectories and conditionally refines the best one. DriveStack-VLA achieves 91.6 PDMS on NAVSIMv1, 91.0 EPDMS on NAVSIMv2 (with the human penalty filter enabled), and a driving score of 79.49 with a success rate of 56.36\% on the closed-loop Bench2Drive. More visualizations are available on our project page: https://anonymous.4open.science/w/drivestack-vla/.

Related papers

• Teaching Vision-Language-Action Models What to See and Where to Look [40.70324197562915]
  Vision-Language-Action (VLA) models have emerged as a promising paradigm for end-to-end autonomous driving.<n>We propose DriveTeach-VLA, a framework that explicitly teaches VLAs what to see and where to look.
  arXiv  Detail & Related papers  (2026-07-02T03:34:32Z)
• LWDrive: Layer-Wise World-Model-Guided Vision-Language Model Planning for Autonomous Driving [13.964957969409655]
  We develop the Layer-Wise World-Model-Guided Driving framework (LWDrive)<n>LWDrive is a VLM planning framework that refines coarse trajectories through layer-wise world-model guidance.<n> Experiments show that LWDrive achieves a score of 92.0 on the NAVSIM benchmark and 89.6 on NAVSIM-v2.
  arXiv  Detail & Related papers  (2026-06-29T07:17:13Z)
• AgentVLN: Towards Agentic Vision-and-Language Navigation [78.739525400071]
  Vision-and-Language Navigation (VLN) requires an embodied agent to ground complex natural-language instructions into long-horizon navigation in unseen environments.<n>In this paper, we propose AgentVLN, a novel and efficient embodied navigation framework that can be deployed on edge computing platforms.
  arXiv  Detail & Related papers  (2026-03-18T12:43:47Z)
• DrivePI: Spatial-aware 4D MLLM for Unified Autonomous Driving Understanding, Perception, Prediction and Planning [94.62097655403683]
  We propose DrivePI, a novel spatial-aware 4D MLLM that serves as a unified Vision-Language-Action framework.<n>Our method jointly performs spatial understanding, 3D perception, prediction (i.e., occupancy flow), and planning (i.e., action outputs) in parallel.<n>With only a 0.5B Qwen2.5 model as MLLM backbone, DrivePI as a single unified model matches or exceeds both existing VLA models and specialized VA models.
  arXiv  Detail & Related papers  (2025-12-14T18:45:54Z)
• Think Before You Drive: World Model-Inspired Multimodal Grounding for Autonomous Vehicles [34.698147360764104]
  ThinkDeeper is a framework that reasons about future spatial states before making grounding decisions.<n>It ranks #1 on the Talk2Car leaderboard and surpasses state-of-the-art baselines on DrivePilot, MoCAD, and RefCOCO/+/g benchmarks.<n>In addition, we present DrivePilot, a multi-source VG dataset in AD, featuring semantic annotations generated by a Retrieval-Augmented Generation (RAG) and Chain-of-Thought pipeline.
  arXiv  Detail & Related papers  (2025-12-03T05:14:16Z)
• FastDriveVLA: Efficient End-to-End Driving via Plug-and-Play Reconstruction-based Token Pruning [75.80110543049783]
  We propose FastDriveVLA, a reconstruction-based vision token pruning framework for autonomous driving.<n>A novel foreground adversarial-background reconstruction strategy is designed to train ReconPruner for the visual encoder of VLA models.<n>Our approach achieves state-of-the-art results on the nuScenes open-loop planning benchmark across different pruning ratios.
  arXiv  Detail & Related papers  (2025-07-31T07:55:56Z)
• OpenDriveVLA: Towards End-to-end Autonomous Driving with Large Vision Language Action Model [24.90085777003393]
  We present OpenDriveVLA, a Vision-Language Action (VLA) model designed for end-to-end autonomous driving.<n>OpenDriveVLA builds upon open-source pre-trained large Vision-Language Models (VLMs) to generate reliable driving actions.
  arXiv  Detail & Related papers  (2025-03-30T14:45:54Z)
• Enhancing End-to-End Autonomous Driving with Latent World Model [78.22157677787239]
  We propose a novel self-supervised learning approach using the LAtent World model (LAW) for end-to-end driving.<n> LAW predicts future scene features based on current features and ego trajectories.<n>This self-supervised task can be seamlessly integrated into perception-free and perception-based frameworks.
  arXiv  Detail & Related papers  (2024-06-12T17:59:21Z)
• Policy Pre-training for End-to-end Autonomous Driving via Self-supervised Geometric Modeling [96.31941517446859]
  We propose PPGeo (Policy Pre-training via Geometric modeling), an intuitive and straightforward fully self-supervised framework curated for the policy pretraining in visuomotor driving. We aim at learning policy representations as a powerful abstraction by modeling 3D geometric scenes on large-scale unlabeled and uncalibrated YouTube driving videos. In the first stage, the geometric modeling framework generates pose and depth predictions simultaneously, with two consecutive frames as input. In the second stage, the visual encoder learns driving policy representation by predicting the future ego-motion and optimizing with the photometric error based on current visual observation only.
  arXiv  Detail & Related papers  (2023-01-03T08:52:49Z)
• MVLidarNet: Real-Time Multi-Class Scene Understanding for Autonomous Driving Using Multiple Views [60.538802124885414]
  We present Multi-View LidarNet (MVLidarNet), a two-stage deep neural network for multi-class object detection and drivable space segmentation. MVLidarNet is able to detect and classify objects while simultaneously determining the drivable space using a single LiDAR scan as input. We show results on both KITTI and a much larger internal dataset, thus demonstrating the method's ability to scale by an order of magnitude.
  arXiv  Detail & Related papers  (2020-06-09T21:28:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.