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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Designing a single model to address multiple tasks has been a long-standingobjective in artificial intelligence. Recently, large language models havedemonstrated exceptional capability in solving different tasks within thelanguage domain. However, a unified model for various graph tasks remainsunderexplored, primarily due to the challenges unique to the graph learningdomain. First, graph data from different areas carry distinct attributes andfollow different distributions. Such discrepancy makes it hard to representgraphs in a single representation space. Second, tasks on graphs diversify intonode, link, and graph tasks, requiring distinct embedding strategies. Finally,an appropriate graph prompting paradigm for in-context learning is unclear. Wepropose \textbf{One for All (OFA)}, the first general framework that can use asingle graph model to address the above challenges. Specifically, OFA proposestext-attributed graphs to unify different graph data by describing nodes andedges with natural language and uses language models to encode the diverse andpossibly cross-domain text attributes to feature vectors in the same embeddingspace. Furthermore, OFA introduces the concept of nodes-of-interest tostandardize different tasks with a single task representation. For in-contextlearning on graphs, OFA introduces a novel graph prompting paradigm thatappends prompting substructures to the input graph, which enables it to addressvaried tasks without fine-tuning. We train the OFA model using graph data frommultiple domains (including citation networks, molecular graphs, knowledgegraphs, etc.) simultaneously and evaluate its ability in supervised, few-shot,and zero-shot learning scenarios. OFA performs well across different tasks,making it the first general-purpose across-domains classification model ongraphs.

One for All: Towards Training One Graph Model for All Classification Tasks