DialogGSR: Generative Subgraph Retrieval for Knowledge Graph-Grounded Dialog Generation

Jinyoung Park, Minseok Joo, Joo-Kyung Kim, Hyunwoo J. Kim.

Official PyTorch implementation of the "DialogGSR: Generative Subgraph Retrieval for Knowledge Graph-Grounded Dialog Generation". (EMNLP 2024)

Enviroment

To install requirements, run:

git clone https://github.com/mlvlab/DialogGSR.git
cd DialogGSR
conda create -n dialoggsr python==3.9
conda activate dialoggsr
sh setting.sh

---

Preparation

OpenDialKG

We utilized OpenDialKG dataset from the link repository. Place the downloaded dataset in the data/ folder.

---

Training

python main.py --output_dir <output_dir>

Contact

If you have any questions, please create an issue on this repository or contact at [email protected].

Citation

If you find our work interesting, please consider giving a ⭐ and citation.

@inproceedings{park2024generative,
  title={Generative Subgraph Retrieval for Knowledge Graph-Grounded Dialog Generation},
  author={Park, Jinyoung and Joo, Minseok and Kim, Joo-Kyung and Kim, Hyunwoo J},
  booktitle={EMNLP},
  year={2024}
}