Probing Multimodal LLMs as World Models for Driving

Open in new window