---
license: cc-by-4.0
pretty_name: 'Diverse Extended Object Nav Episode Dataset for HSSD Scenes '
---
<div align="center">
<h1>Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning</h1>
<p>
  <a href="https://gunshigupta.netlify.app/">Gunshi Gupta</a>,
  <a href="https://www.karmeshyadav.com/">Karmesh Yadav</a>,
  <a href="https://faculty.cc.gatech.edu/~zk15/">Zsolt Kira</a>,
  <a href="https://www.cs.ox.ac.uk/people/yarin.gal/website/">Yarin Gal</a>,
  <a href="https://www.linkedin.com/in/rahaf-aljundi/">Rahaf Aljundi</a>
  
</p>
<p>
  <a href="https://arxiv.org/pdf/2510.19732">Paper Link</a>&ensp;
  <a href="https://github.com/gunshi/memo/">Code</a>
</p>
</div>

This is the episode dataset associated with the paper titled, Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning. Please checkout our paper and codebase for more details. If you find this work useful, please cite:
```bibtex
@article{gupta2025memo,
  title={Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning},
  author={Gupta, Gunshi and Yadav, Karmesh and Kira, Zsolt and Gal, Yarin and Aljundi, Rahaf},
  journal={arXiv preprint arXiv:2510.19732},
  year={2025}
}
```