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

Instead of relying on human-annotated training samples to build a classifier,weakly supervised scientific paper classification aims to classify papers onlyusing category descriptions (e.g., category names, category-indicativekeywords). Existing studies on weakly supervised paper classification are lessconcerned with two challenges: (1) Papers should be classified into not onlycoarse-grained research topics but also fine-grained themes, and potentiallyinto multiple themes, given a large and fine-grained label space; and (2) fulltext should be utilized to complement the paper title and abstract forclassification. Moreover, instead of viewing the entire paper as a long linearsequence, one should exploit the structural information such as citation linksacross papers and the hierarchy of sections and paragraphs in each paper. Totackle these challenges, in this study, we propose FUTEX, a framework that usesthe cross-paper network structure and the in-paper hierarchy structure toclassify full-text scientific papers under weak supervision. A network-awarecontrastive fine-tuning module and a hierarchy-aware aggregation module aredesigned to leverage the two types of structural signals, respectively.Experiments on two benchmark datasets demonstrate that FUTEX significantlyoutperforms competitive baselines and is on par with fully supervisedclassifiers that use 1,000 to 60,000 ground-truth training samples.

Weakly Supervised Multi-Label Classification of Full-Text Scientific Papers