Metraipolis › archive › dc_be372fe789e9

OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher — arXiv:2609.20756

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.20756
License (as archived)
CC0-1.0
Archived
2026-10-10T04:09:30.417Z
SHA-256 of the archived text
15bad5c55acbfce2d9e589935086f4e8e0e5f373057477daa45c49d4a340d6bf
Stable URL of this page
https://metraipolis.com/documento/dc_be372fe789e9

How to cite

OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher — arXiv:2609.20756 — arXiv (en), https://arxiv.org/abs/2609.20756 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_be372fe789e9 (Metraipolis archive dc_be372fe789e9, sha256 15bad5c55acbfce2d9e589935086f4e8e0e5f373057477daa45c49d4a340d6bf)
BibTeX
@misc{dcbe372fe789e9,
  title = {OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher — arXiv:2609.20756},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20756},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_be372fe789e9; sha256 15bad5c55acbfce2d9e589935086f4e8e0e5f373057477daa45c49d4a340d6bf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20756. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As scaling pre-training data alone yields diminishing returns, post-training is becoming increasingly important across physical AI domains such as autonomous driving. End-to-end driving policies are pre-trained in open loop with behavior cloning on human demonstrations. However, compounding errors during closed-loop deployment can take the vehicle outside the training data distribution, increasing the risk of safety-critical incidents. Closed-loop post-training can mitigate this risk but requires costly simulation for sensor-based policies. We propose OPTED (on-policy fine-tuning for end-to-end driving) which decouples reinforcement learning from the post-training of the end-to-end policy: a privileged teacher is trained using RL on vectorized inputs (HD-map and bounding boxes). This teacher then provides supervision to the pre-trained student during closed-loop post-training. We apply OPTED to two camera-based models, TransFuser and VaVAM, and fine-tune them in AlpaSim, using neural reconstructions (3DGS) of real driving logs. Driving scores increase by factors of 1.6$\times$ and 9.5$\times$, respectively. In controlled experiments OPTED matches closed-loop performance with approximately three orders of magnitude fewer simulator interactions than direct RL post-training, while staying closer to the human prior. Project page: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.20756 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T03:48:18.479Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).