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

Node classification is a classical graph representation learning task onwhich Graph Neural Networks (GNNs) have recently achieved strong results.However, it is often believed that standard GNNs only work well for homophilousgraphs, i.e., graphs where edges tend to connect nodes of the same class.Graphs without this property are called heterophilous, and it is typicallyassumed that specialized methods are required to achieve strong performance onsuch graphs. In this work, we challenge this assumption. First, we show thatthe standard datasets used for evaluating heterophily-specific models haveserious drawbacks, making results obtained by using them unreliable. The mostsignificant of these drawbacks is the presence of a large number of duplicatenodes in the datsets Squirrel and Chameleon, which leads to train-test dataleakage. We show that removing duplicate nodes strongly affects GNN performanceon these datasets. Then, we propose a set of heterophilous graphs of varyingproperties that we believe can serve as a better benchmark for evaluating theperformance of GNNs under heterophily. We show that standard GNNs achievestrong results on these heterophilous graphs, almost always outperformingspecialized models. Our datasets and the code for reproducing our experimentsare available at https://github.com/yandex-research/heterophilous-graphs

A critical look at the evaluation of GNNs under heterophily: are we really making progress?