JJYDXFS/RAMP_pretrained_data
收藏Hugging Face2026-03-26 更新2026-03-29 收录
下载链接：
https://hf-mirror.com/datasets/JJYDXFS/RAMP_pretrained_data
下载链接
链接失效反馈官方服务：
资源简介：
---
license: apache-2.0
---
# Dataset Details
This dataset contains pretrained data constructed from academic text-rich graphs ([MAPLE](https://github.com/yuzhimanhua/MAPLE)) spanning three disciplines: Economics, Mathematics, and Geology. It is used for pretraining [RAMP](https://arxiv.org/abs/2603.14937) (Raw-text Anchored Message Passing), which recasts the LLM as a graph-native aggregation operator on text-rich graphs.
The dataset includes the following files:
- `pretrained_Economics.json`
- `pretrained_Mathematics.json`
- `pretrained_Geology.json`
This is a release from our paper [LLM as Graph Kernel: Rethinking Message Passing on Text-Rich Graphs](https://arxiv.org/abs/2603.14937), so please cite it if using this dataset.
# Citation
```bibtex
@article{zhang2026llm,
title={LLM as Graph Kernel: Rethinking Message Passing on Text-Rich Graphs},
author={Zhang, Ying and Yu, Hang and Zhang, Haipeng and Di, Peng},
journal={arXiv preprint arXiv:2603.14937},
year={2026}
}
```
[](https://arxiv.org/abs/2603.14937)
提供机构：
JJYDXFS
