ConGraT: Self-Supervised Contrastive Pretraining for Joint Graph and Text Embeddings

1 code implementation • 23 May 2023 • William Brannon, Suyash Fulay, Hang Jiang, Wonjune Kang, Brandon Roy, Jad Kabbara, Deb Roy

We propose ConGraT(Contrastive Graph-Text pretraining), a general, self-supervised method for jointly learning separate representations of texts and nodes in a parent (or ``supervening'') graph, where each text is associated with one of the nodes.

Contrastive Learning Link Prediction

CommunityLM: Probing Partisan Worldviews from Language Models

1 code implementation • COLING 2022 • Hang Jiang, Doug Beeferman, Brandon Roy, Deb Roy

As political attitudes have diverged ideologically in the United States, political speech has diverged lingusitically.