---
license: mit
pipeline_tag: reinforcement-learning
---
<p align="left"> 
  <a href="https://github.com/qortmdgh4141/GAS">
  <img alt="Code" src="https://img.shields.io/badge/Code--lightgrey?style=social&logo=github"> </a> 
  
  <a href="https://arxiv.org/abs/2506.07744"> 
  <img alt="Paper" src="https://img.shields.io/badge/Paper--lightgrey?style=social&logo=arxiv"> </a> 
  
  <a href="https://qortmdgh4141.github.io/projects/GAS/"> 
  <img alt="Project Page" src="https://img.shields.io/badge/Project page--lightgrey?style=social&logo=rocket"> </a> 
  
  <a href="https://www.youtube.com/watch?v=6mxRlbn2_6s"> 
  <img alt="YouTube Video Views" src="https://img.shields.io/badge/Talk (10min)--lightgrey?style=social&logo=youtube"> </a> 
</p>

This repository provides official checkpoints for **[Graph-Assisted Stitching](https://arxiv.org/abs/2506.07744)** (**GAS**).

See the [GitHub repository](https://github.com/qortmdgh4141/GAS) and [Project page](https://qortmdgh4141.github.io/projects/GAS/) for more details.

## Citation
```bibtex
@inproceedings{gas_baek2025,
    title={Graph-Assisted Stitching for Offline Hierarchical Reinforcement Learning},
    author={Seungho Baek and Taegeon Park and Jongchan Park and Seungjun Oh and Yusung Kim},
    booktitle={International Conference on Machine Learning (ICML)},
    year={2025},
}
```