# FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model

Source: [FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model](<https://www.sota2.com/research/paper/693f961b6cfd488761a81c7a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-16T05:02:13.709Z

Authors: Hongbin Lin, Yiming Yang, Yifan Zhang, Chaoda Zheng, Jie Feng, Sheng Wang, Zhennan Wang, Shijia Chen, Boyang Wang, Yu Zhang, Xianming Liu, Shuguang Cui, Zhen Li

Published: 2025-12-12T02:12:49.000Z

## Abstract

In autonomous driving, end-to-end planners learn scene representations from raw sensor data and utilize them to generate a motion plan or control actions. However, exclusive reliance on the current scene for motion planning may result in suboptimal responses in highly dynamic traffic environments where ego actions further alter the future scene. To model the evolution of future scenes, we leverage the World Model to represent how the ego vehicle and its environment interact and change over time, which entails complex reasoning. The Chain of Thought (CoT) offers a promising solution by forecasting a sequence of future thoughts that subsequently guide trajectory refinement. In this paper, we propose FutureX, a CoT-driven pipeline that enhances end-to-end planners to perform complex motion planning via future scene latent reasoning and trajectory refinement. Specifically, the Auto-think Switch examines the current scene and decides whether additional reasoning is required to yield a higher-quality motion plan. Once FutureX enters the Thinking mode, the Latent World Model conducts a CoT-guided rollout to predict future scene representation, enabling the Summarizer Module to further refine the motion plan. Otherwise, FutureX operates in an Instant mode to generate motion plans in a forward pass for relatively simple scenes. Extensive experiments demonstrate that FutureX enhances existing methods by producing more rational motion plans and fewer collisions without compromising efficiency, thereby achieving substantial overall performance gains, e.g., 6.2 PDMS improvement for TransFuser on NAVSIM. Code will be released.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [End-to-End Autonomous Driving on NAVSIM (Navtest)](<https://www.sota2.com/research/sota/end-to-end-autonomous-driving-on-navsim-navtest.md>): End-to-end Autonomous Driving; NAVSIM (navtest); NC: 99.6.

- [Autonomous Driving on CARLA Longest6 Town1](<https://www.sota2.com/research/sota/autonomous-driving-on-carla-longest6-town1.md>): Autonomous Driving; CARLA Longest6 Town1; DS: 84.3.

- [Autonomous Driving on CARLA Longest6 Town5](<https://www.sota2.com/research/sota/autonomous-driving-on-carla-longest6-town5.md>): Autonomous Driving; CARLA Longest6 Town5; Driving Score (DS): 53.7.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.11226>)

[PDF](<https://arxiv.org/pdf/2512.11226.pdf>)
