CoReVLA: A Dual-Stage End-to-End Autonomous Driving Framework for Long-Tail Scenarios via Collect-and-Refine

This repository contains the model weights of CoReVLA, a dual-stage end-to-end autonomous driving framework designed to handle long-tail scenarios via a collect-and-refine strategy.

• 📂 Code: https://github.com/FanGShiYuu/CoReVLA
• 📌 Important Notice: We discovered that the previously uploaded model was incomplete. Please use models uploaded after 10/12/2025 as the correct version.
• 📧 Contact: fangshiyu@tongji.edu.cn for any questions or issues.