SparseDriveV2 is an end-to-end multi-modal planning framework for autonomous driving. It demonstrates that performance consistently improves as trajectory anchors become denser, achieving state-of-the-art results through a scalable vocabulary representation and a factorized scoring strategy.
SparseDriveV2 pushes the performance boundary of scoring-based planning through two complementary innovations:
This approach allows the model to scale its trajectory vocabulary to be 32x denser than prior methods while maintaining computational efficiency.
The model achieves state-of-the-art performance using a lightweight ResNet-34 backbone:
1@article{sun2026sparsedrivev2,
2 title={SparseDriveV2: Scoring is All You Need for End-to-End Autonomous Driving},
3 author={Sun, Wenchao and Lin, Xuewu and Chen, Keyu and Pei, Zixiang and Li, Xiang and Shi, Yining and Zheng, Sifa},
4 journal={arXiv preprint arXiv:2603.29163},
5 year={2026}
6}
