ln2697/TFv6

LEAD: Minimizing Learner–Expert Asymmetry in End-to-End Driving

Project Page | Paper | Code

Official model weights for TransFuser v6 (TFv6), a set of CARLA driving policy checkpoints accompanies our paper LEAD: Minimizing Learner–Expert Asymmetry in End-to-End Driving.

We release the complete pipeline (covering scenario descriptions, expert driver, data preprocessing scripts, training code, and evaluation infrastructure) required to achieve state-of-the-art closed-loop performance on the Bench2Drive benchmark. Built around the CARLA simulator, the stack features a data-centric design with:

• Extensive visualization suite and runtime type validation for easier debugging.
• Optimized storage format, packs 72 hours of driving in ~200GB.
• Native support for NAVSIM and Waymo Vision-based E2E, with LEAD extending these benchmarks through closed-loop simulation and synthetic data for additional supervision during training

Find more information on https://github.com/autonomousvision/lead.

Usage

For setup instructions, data collection, and evaluation scripts, please refer to the official GitHub repository and the full documentation.

Citation

If you find this work useful, please cite:

@article{Nguyen2025ARXIV,
  title={LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving},
  author={Nguyen, Long and Fauth, Micha and Jaeger, Bernhard and Dauner, Daniel and Igl, Maximilian and Geiger, Andreas and Chitta, Kashyap},
  journal={arXiv preprint arXiv:2512.20563},
  year={2025}
}

License

This project is released under the MIT License