multimodalfactneutralThe automotive industry is increasingly exploring vision-language models to interpret camera-recorded in-car scenes for safety and assistance applicationsComputer Vision27 Jul 2026http://arxiv.org/abs/2607.02300v1