Fetching the paper…

Video Token Sparsification for Efficient Multimodal LLMs in Autonomous Driving · Around