A ​ u ​ t ​ o ​ D ​ r ​ i ​ v ​ e ​ - ​ P 3 AutoDrive\text{-}P^{3} : Unified Chain of Perception–Prediction–Planning Thought via Reinforcement Fine-Tuning

Yuqi Ye, Zijian Zhang, Junhong Lin, Shangkun Sun, Changhao Peng, Wei Gao · 2026

Read A ​ u ​ t ​ o ​ D ​ r ​ i ​ v ​ e ​ - ​ P 3 AutoDrive\text{-}P^{3} : Unified Chain of Perception–Prediction–Planning Thought via Reinforcement Fine-Tuning by Yuqi Ye, Zijian Zhang, Junhong Lin, Shangkun Sun, Changhao Peng, Wei Gao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-28116.txt.

Description

Ye and five co-authors present AutoDrive-P3, a driving system that reasons through perception, prediction, and planning as one chain. Many vision-language drivers either skip reasoning or split the three tasks into silos. AutoDrive-P3 trains on a new chain-of-thought dataset and a staged reinforcement method that rewards each step in turn. It also offers a fast mode for quick calls. On open-loop and closed-loop benchmarks, it reached top planning scores.

Catalog record

CREATORYuqi Ye, Zijian Zhang, Junhong Lin, Shangkun Sun, Changhao Peng, Wei Gao
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69513
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.28116

Browse the full library · About the Tower of Babel