Drive My Way: Preference Alignment of Vision-Language-Action Model for Personalized Driving

Zehao Wang, Huaide Jiang, Shuaiwu Dong, Yuping Wang, Hang Qiu et al.|March 26, 2026arXiv

Key Takeaway

Autonomous driving systems can be personalized to match individual driver styles by learning user embeddings from driving data and conditioning the driving policy on these embeddings, enabling more human-centered autonomous vehicles.

Summary

This paper presents Drive My Way, a personalized autonomous driving system that learns individual driver preferences and adapts to real-time instructions.

multimodal agents applications

Key Terms

vision-language-action-model user-embedding preference-alignment end-to-end-driving