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

返回详情 VLA / Vision-Language-Action 每日论文卡