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

Collaborative filtering (CF), as a standard method for recommendation withimplicit feedback, tackles a semi-supervised learning problem where mostinteraction data are unobserved. Such a nature makes existing approaches highlyrely on mining negatives for providing correct training signals. However,mining proper negatives is not a free lunch, encountering with a trickytrade-off between mining informative hard negatives and avoiding false ones. Wedevise a new approach named as Hardness-Aware Debiased ContrastiveCollaborative Filtering (HDCCF) to resolve the dilemma. It could sufficientlyexplore hard negatives from two-fold aspects: 1) adaptively sharpening thegradients of harder instances through a set-wise objective, and 2) implicitlyleveraging item/user frequency information with a new sampling strategy. Tocircumvent false negatives, we develop a principled approach to improve thereliability of negative instances and prove that the objective is an unbiasedestimation of sampling from the true negative distribution. Extensiveexperiments demonstrate the superiority of the proposed model over existing CFmodels and hard negative mining methods.

Trading Hard Negatives and True Negatives: A Debiased Contrastive Collaborative Filtering Approach