This is the dataset for GraphMERT: Efficient and Scalable Distillation of Reliable Knowledge Graphs from Unstructured Data.
This folder contains the training data for GraphMert model.
If you use GraphMERT (code, models, data or data processing scripts) in your work, please cite our TMLR paper:
@article{
belova2026graphmert,
title={Graph{MERT}: {E}fficient and Scalable Distillation of Reliable Knowledge Graphs from Unstructured Data},
author={Margarita Belova and Jiaxin Xiao and Shikhar Tuli and Niraj Jha},
journal={Transactions on Machine Learning Research},
issn={2835-8856},
year={2026},
url={https://openreview.net/forum?id=tnXSdDhvqc},
}