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

Industry recommender systems usually suffer from highly-skewed long-tail itemdistributions where a small fraction of the items receives most of the userfeedback. This skew hurts recommender quality especially for the item sliceswithout much user feedback. While there have been many research advances madein academia, deploying these methods in production is very difficult and veryfew improvements have been made in industry. One challenge is that thesemethods often hurt overall performance; additionally, they could be complex andexpensive to train and serve. In this work, we aim to improve tail itemrecommendations while maintaining the overall performance with less trainingand serving cost. We first find that the predictions of user preferences arebiased under long-tail distributions. The bias comes from the differencesbetween training and serving data in two perspectives: 1) the itemdistributions, and 2) user's preference given an item. Most existing methodsmainly attempt to reduce the bias from the item distribution perspective,ignoring the discrepancy from user preference given an item. This leads to asevere forgetting issue and results in sub-optimal performance.  To address the problem, we design a novel Cross Decoupling Network (CDN) (i)decouples the learning process of memorization and generalization on the itemside through a mixture-of-expert architecture; (ii) decouples the user samplesfrom different distributions through a regularized bilateral branch network.Finally, a new adapter is introduced to aggregate the decoupled vectors, andsoftly shift the training attention to tail items. Extensive experimentalresults show that CDN significantly outperforms state-of-the-art approaches onbenchmark datasets. We also demonstrate its effectiveness by a case study ofCDN in a large-scale recommendation system at Google.

Empowering Long-tail Item Recommendation through Cross Decoupling Network (CDN)