XPeng shares autonomous driving AI roadmap at CVPR 2026
XPeng shared its World Model roadmap at the CVPR 2026 Workshop, highlighting proactive reasoning, controllable generation, and long-horizon forecasting as key advancements for autonomous driving.

*this image is generated using AI for illustrative purposes only.
XPeng shared its World Model roadmap at the CVPR 2026 Workshop on Foundation Model Deployment for Embodied Intelligence. Xianming Liu, Head of XPeng Group's General Intelligence Center, unveiled the roadmap, highlighting proactive reasoning, controllable generation, and long-horizon forecasting as key capabilities for next-generation autonomous driving.
Key Capabilities
The roadmap focuses on advancing autonomous driving through several core features:
- Proactive reasoning: Enhancing decision-making capabilities in real-time scenarios.
- Controllable generation: Allowing precise control over AI-generated outputs.
- Long-horizon forecasting: Improving the ability to predict future events over extended periods.
These capabilities aim to strengthen XPeng's position in the autonomous driving sector by leveraging advanced AI models.
How will XPeng's World Model roadmap impact its competitive position against other autonomous driving companies?
What are the potential challenges in implementing proactive reasoning and long-horizon forecasting in real-world scenarios?
How might these advancements influence XPeng's partnerships with tech firms or automakers?

























