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

Partition-based methods are increasingly-used in extreme multi-labelclassification (XMC) problems due to their scalability to large output spaces(e.g., millions or more). However, existing methods partition the large labelspace into mutually exclusive clusters, which is sub-optimal when labels havemulti-modality and rich semantics. For instance, the label "Apple" can be thefruit or the brand name, which leads to the following research question: can wedisentangle these multi-modal labels with non-exclusive clustering tailored fordownstream XMC tasks? In this paper, we show that the label assignment problemin partition-based XMC can be formulated as an optimization problem, with theobjective of maximizing precision rates. This leads to an efficient algorithmto form flexible and overlapped label clusters, and a method that canalternatively optimizes the cluster assignments and the model parameters forpartition-based XMC. Experimental results on synthetic and real datasets showthat our method can successfully disentangle multi-modal labels, leading tostate-of-the-art (SOTA) results on four XMC benchmarks.

Label Disentanglement in Partition-based Extreme Multilabel Classification