LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models

Code for LLM-Planner.

Check project website for an overview and a demo.

What's Here

• hlp/: A high-level prompt generator and kNN dataset from our paper. Just bring your low-level controller (and an LLM)!
• e2e/: A end-to-end agent that uses a LLM-Planner.

Quickstart

Check out the hlp/ and e2e/ README for more details.

Implementation Examples

We provide examples of how the community has been using our work. We appreciate everyone's interest!

• DEDER – ICML 2024
• ReALFRED – ECCV 2024
• ExRAP – NeurIPS 2024
• FLARE – AAAI 2025
• NeSyC – ICLR 2025
• Socratic Planner – ICRA 2025

Acknowledgements

We thank OSUNLP for providing valuable feedback and suggestions.

License

• LLM-Planner - MIT License

Contact

Questions or issues? File an issue or contact Luke Song

Citation Information

@InProceedings{song2023llmplanner,
  author    = {Song, Chan Hee and Wu, Jiaman and Washington, Clayton and Sadler, Brian M. and Chao, Wei-Lun and Su, Yu},
  title     = {LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  month     = {October},
  year      = {2023},
}