HERMES introduces the first unified Driving World Model that seamlessly integrates 3D scene understanding with future scene generation for autonomous vehicles.

• Combines scene interpretation and predictive capabilities in one framework
• Enables vehicles to simultaneously understand current road conditions and predict future scenarios
• Overcomes limitations of existing models that focus on generation without comprehension
• Advances both engineering capability and safety aspects of autonomous driving systems

This breakthrough has significant implications for developing more intelligent, contextually aware self-driving systems that can better navigate complex real-world environments.

HERMES: A Unified Self-Driving World Model for Simultaneous 3D Scene Understanding and Generation