cs.RONov 14, 2026

SuperGlance: Driving at a Glance with a Vision-Language Model

Authors: Jieyuan Pei, Xin Yan, Mingwei Xu, Haiyi Zhang, Siyu Zou, Zhangyi Hu, Zhuoxuan Li, Yuguang Yang, 

Organizations: Institute for AI Industry Research (AIR), Tsinghua University · HiThink Research · The Hong Kong University of Science and Technology (Guangzhou) · Agency for Science, Technology and Research (A*STAR), Singapore · 

Abstract

Autonomous driving is turning to vision-language-action models (VLAs). They answer and plan by decoding, one step at a time. Each step is another forward pass, so a plan can take over a second of GPU time. Embodied AI often pairs such a slow, reasoning System 2 with a fast System 1 that acts. Human drivers plan at a glance, with System 1. Faster designs usually put System 1 outside the vision-language model (VLM), in a planner without its abilities. Yet before a VLM writes its first token, it has already read the scene into one hidden state. We show that, once trained, this hidden state is enough to drive and to answer questions about what the model sees. SuperGlance reads it to turn a decoder-only VLM into an encoder-style driving policy. Its linear head scores a fixed set of 5,273 candidate trajectories. With the set fixed, rewards are computed offline, so supervised targets can be reward-weighted and reinforcement learning maximizes the exact expected reward. Before and during policy training, multiple-choice training keeps the VLM's visual abilities in one pass, so the policy still answers questions about driving scenes. Training questions are built against shortcuts. Each must be answered right in two option orders. Against drift, a KL term keeps each stage near its starting model. One pass offers VLAs a more plausible path to System 1. SuperGlance answers visual multiple-choice questions as well as its base VLM does by decoding. It plans in 84 ms of GPU time, six times faster than Qwen-Drive-1.0, and at least as well on NAVSIM, PAI-AV and WOD-E2E. In AlpaSim closed loop, km per at-fault incident holds for both planners up to a 300-ms delay and drops beyond. With GPU time charged as delay, SuperGlance's lead over Qwen-Drive-1.0 grows from 2.8 to 3.9 times. Delayed by 1.2 s, our model falls below Qwen-Drive-1.0 at its own 518 ms.

Figures & tables

Appendix figures & tables14 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Teaching Vision-Language-Action Models What to See and Where to Look

    Jul 2, 2026Yuguang Yang, Canyu Chen, Zhewen Tan +10End-to-End Autonomous DrivingVLM Distillation

  2. FlashDrive: Flash Vision-Language-Action Inference for Autonomous Driving

    Aug 13, 2026Zekai Li, Yihao Liang, Hongfei Zhang +3Efficient VLA Model InferenceAutonomous Driving

  3. DriveMA: Driving Vision-Language-Action Models with verifiable Meta-Actions

    May 29, 2026Weicheng Zheng, Yixin Huang, Qiao Sun +2VLMs for Autonomous DrivingAutonomous Driving