Pretrained Language Models to Solve Graph Tasks in Natural Language

Abstract

Pretrained large language models (LLMs) are powerful learners in a variety of language tasks. We explore if LLMs can learn from graph-structured data when the graphs are described using natural language. We explore data augmentation and pretraining specific to the graph domain and show that LLMs such as GPT-2 and GPT-3 are promising alternatives to graph neural networks.

Cite

Text

Wenkel et al. "Pretrained Language Models to Solve Graph Tasks in Natural Language." ICML 2023 Workshops: SPIGM, 2023.

Markdown

[Wenkel et al. "Pretrained Language Models to Solve Graph Tasks in Natural Language." ICML 2023 Workshops: SPIGM, 2023.](https://mlanthology.org/icmlw/2023/wenkel2023icmlw-pretrained/)

BibTeX

@inproceedings{wenkel2023icmlw-pretrained,
  title     = {{Pretrained Language Models to Solve Graph Tasks in Natural Language}},
  author    = {Wenkel, Frederik and Wolf, Guy and Knyazev, Boris},
  booktitle = {ICML 2023 Workshops: SPIGM},
  year      = {2023},
  url       = {https://mlanthology.org/icmlw/2023/wenkel2023icmlw-pretrained/}
}