DriveVLM-RL: Neuroscience-Inspired Reinforcement Learning with Vision-Language Models for Safe and Deployable Autonomous Driving

ArXi:2603.18315v1 Announce Type: cross Ensuring safe decision-making in autonomous vehicles remains a fundamental challenge despite rapid advances in end-to-end learning approaches. Traditional reinforcement learning (RL) methods rely on manually engineered rewards or sparse collision signals, which fail to capture the rich contextual understanding required for safe driving and make unsafe exploration unavoidable in real-world settings.