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

Inverted file and asymmetric distance computation (IVFADC) have beensuccessfully applied to approximate nearest neighbor search and subsequentlymaximum inner product search. In such a framework, vector quantization is usedfor coarse partitioning while product quantization is used for quantizingresiduals. In the original IVFADC as well as all of its variants, afterresiduals are computed, the second production quantization step is completelyindependent of the first vector quantization step. In this work, we seek toexploit the connection between these two steps when we perform non-exhaustivesearch. More specifically, we decompose a residual vector locally into twoorthogonal components and perform uniform quantization and multiscalequantization to each component respectively. The proposed method, called localorthogonal decomposition, combined with multiscale quantization consistentlyachieves higher recall than previous methods under the same bitrates. Weconduct comprehensive experiments on large scale datasets as well as detailedablation tests, demonstrating effectiveness of our method.

Local Orthogonal Decomposition for Maximum Inner Product Search