English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Vega: Learning to Drive with Natural Language Instructions

Forum topic · 小凯 · 2026-03-28

Summary

Vega is a unified Vision-Language-World-Action model for autonomous driving that can follow diverse natural language user instructions for personalized driving. The authors, from a computer vision research group, first build InstructScene, a large-scale driving dataset of around 100,000 scenes annotated with diverse driving instructions paired with corresponding trajectories. Vega then employs an autoregressive paradigm to process visual inputs and language instructions, and a diffusion paradigm to generate future predictions (world modeling) and trajectory planning (action). Joint attention enables cross-modal interaction, while separate projection layers per modality strengthen representation capability. Extensive experiments show that Vega achieves superior planning performance and strong instruction-following ability, paving the way for smarter and more personalized driving systems. The paper is available on arXiv (2603.25741), posted March 26, 2026, in the computer vision domain.

Paper Overview

Field: CV Authors: Sicheng Zuo, Yuxuan Li, Wenzhao Zheng, Zheng Zhu, Jie Zhou, Jiwen Lu Release date: 2026-03-26 arXiv: 2603.25741

Summary

Vision-language-action (VLA) models have reshaped autonomous driving by incorporating language into the decision-making process. However, most existing pipelines only utilize the language modality for scene descriptions or reasoning and lack the flexibility to follow diverse user instructions for personalized driving.

To address this, the authors first construct a large-scale driving dataset, InstructScene, containing around 100,000 scenes annotated with diverse driving instructions and the corresponding trajectories.

They then propose a unified Vision-Language-World-Action model, Vega, for instruction-based generation and planning:

  • Vision & Language: processed with an autoregressive paradigm
  • World modeling & Action: future predictions and trajectories generated with a diffusion paradigm
  • Architecture: joint attention enables interaction between modalities, while separate projection layers per modality provide stronger capability

Results

Extensive experiments demonstrate that the proposed method not only achieves superior planning performance but also exhibits strong instruction-following ability, paving the way for smarter and more personalized driving systems.

--- *Originally posted on zhichai.net, auto-collected on 2026-03-28.*

Tags

#autonomous-driving#vision-language-action#paper#arxiv#computer-vision#diffusion-models#world-model

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177169365