English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Graph property prediction tasks are important and numerous. While each taskoffers a small size of labeled examples, unlabeled graphs have been collectedfrom various sources and at a large scale. A conventional approach is traininga model with the unlabeled graphs on self-supervised tasks and then fine-tuningthe model on the prediction tasks. However, the self-supervised task knowledgecould not be aligned or sometimes conflicted with what the predictions needed.In this paper, we propose to extract the knowledge underlying the large set ofunlabeled graphs as a specific set of useful data points to augment eachproperty prediction model. We use a diffusion model to fully utilize theunlabeled graphs and design two new objectives to guide the model's denoisingprocess with each task's labeled data to generate task-specific graph examplesand their labels. Experiments demonstrate that our data-centric approachperforms significantly better than fifteen existing various methods on fifteentasks. The performance improvement brought by unlabeled data is visible as thegenerated labeled examples unlike the self-supervised learning.

Data-Centric Learning from Unlabeled Graphs with Diffusion Model