Evaluation of Safety Cognition Capability in Vision-Language Models for Autonomous Driving

A new framework called SCD-Bench has been introduced to evaluate the safety cognition capabilities of vision-language models in autonomous driving. This is significant because ensuring safety in these systems is crucial, especially as current research has mainly focused on traditional benchmarks. By addressing safety in interactive driving scenarios, this framework aims to enhance the reliability of autonomous vehicles, making them safer for everyone on the road.