---
license: apache-2.0
---
# CoIRL-AD: Collaborative–Competitive Imitation–Reinforcement Learning in Latent World Models for Autonomous Driving

<div align="center">
<a href="https://seu-zxj.github.io/">Xiaoji Zheng</a>*, 
<a href="https://ziyuan-yang.github.io">Yangzi Yuan</a>*, 
<a href="https://github.com/Ian-cyh">Yanhao Chen</a>, 
<a href="https://github.com/doraemonaaaa">Yuhang Peng</a>, 
<a href="https://github.com/TANGXTONG1">Yuanrong Tang</a>, 
<a href="https://openreview.net/profile?id=~Gengyuan_Liu1">Gengyuan Liu</a>,
<a href="https://scholar.google.com/citations?user=_Wrx_yEAAAAJ">Bokui Chen</a>‡ and 
<a href="https://scholar.google.com/citations?user=AktmI14AAAAJ">Jiangtao Gong</a>‡.
<div>
*: Equal contribution.
‡: Corresponding authors.
</div>

<div>
<a href="https://seu-zxj.github.io/CoIRL-AD"><img alt="Static Badge" src="https://img.shields.io/badge/_-page-blue?style=flat&logo=githubpages&logoColor=white&logoSize=auto&labelColor=gray"></a>
<a href="https://arxiv.org/abs/2510.12560"><img alt="Static Badge" src="https://img.shields.io/badge/arxiv-paper-red?logo=arxiv"></a>
<a href="https://github.com/SEU-zxj/CoIRL-AD"><img alt="Static Badge" src="https://img.shields.io/badge/github-code-white?logo=github"></a>
<a href="https://huggingface.co/Student-Xiaoji/CoIRL-AD-models"><img alt="Static Badge" src="https://img.shields.io/badge/hf-models-yellow?logo=huggingface"></a>
</div>

</div>

**CoIRL-AD** introduces a dual-policy framework that unifies imitation learning (IL) and reinforcement learning (RL) through a collaborative–competitive mechanism within a latent world model.  
The framework enhances generalization and robustness in end-to-end autonomous driving without relying on external simulators.

![main figure](./assets/main_figure.jpg)

Here we provide our model checkpoints (see `/ckpts`), info files (see `/info-files`) for dataloader to download and reproduce our experiment results.

## For more details, please refer to our [github repo](https://github.com/SEU-zxj/CoIRL-AD).