SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World

Jungho Kim, Jiyong Oh, Seunghoon Yu, Hongjae Shin, Donghyuk Kwak, Jun Won Choi

Seoul National University, ADR Lab

CVPR 2026 Highlight

🔔 News

• [2026/04]: SafeDrive is awarded as CVPR 2026 Highlight! ⭐
• [2026/03]: We will release the full code & checkpoints of SafeDrive.
• [2026/02]: SafeDrive is accepted at CVPR 2026! 🔥

📽️ Framework

Comparison of end-to-end planning paradigms and the SafeDrive framework. (a) Dense world models provide limited modeling of instance-centric interactions, whereas sparse world models capture them effectively. (b) Scene-level safety evaluation is coarse, while fine-grained evaluation identifies the specific agents and timestamps associated with potential risks. (c) SafeDrive leverages a sparse world model and fine-grained safety reasoning to generate safe trajectories.

⚡ Main Result

📃 Bibtex

If you find this work useful for your research or projects, please consider citing the following BibTeX entry.

@inproceedings{safedrive,
  title={SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World},
  author={Kim, Jungho and Oh, Jiyong and Yu, Seunghoon and Shin, Hongjae and Kwak, Donghyuk and Choi, Jun Won},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2026}
}

🙏 Acknowledgement

This project builds upon several outstanding open-source projects. We gratefully acknowledge the following key contributions.

• NAVSIM, DiffusionDrive, WoTE, BEVFormer, GTRS, iPad