Accelerating Structured Chain-of-Thought in Autonomous Vehicles
Yi Gu Yan Wang Yuxiao Chen Yurong You Wenjie Luo Yue Wang Wenhao Ding Boyi Li Heng Yang Boris Ivanovic Marco Pavone · 2026
Read Accelerating Structured Chain-of-Thought in Autonomous Vehicles by Yi Gu Yan Wang Yuxiao Chen Yurong You Wenjie Luo Yue Wang Wenhao Ding Boyi Li Heng Yang Boris Ivanovic Marco Pavone (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-2602-02864.txt.
Description
Self-driving cars think too slowly when they reason step by step, and Yi Gu and 10 co-authors tackle that delay in 2026. Vision-language-action models use chain-of-thought reasoning, written steps before acting. Writing each step in order creates latency that real-time driving cannot afford.
Their method, FastDriveCoT, maps the reasoning steps into a dependency graph. Steps that do not depend on each other, like spotting key objects and summarizing traffic rules, run in parallel. Other steps still follow in order. Experiments across several model designs follow. Reasoning ran three to four times faster, and total delay fell sharply. The driving benefits of the reasoning itself were preserved.
Catalog record
| CREATOR | Yi Gu Yan Wang Yuxiao Chen Yurong You Wenjie Luo Yue Wang Wenhao Ding Boyi Li Heng Yang Boris Ivanovic Marco Pavone |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-02864