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

Classical information retrieval (IR) methods, such as query likelihood andBM25, score documents independently w.r.t. each query term, and then accumulatethe scores. Assuming query term independence allows precomputing term-documentscores using these models---which can be combined with specialized datastructures, such as inverted index, for efficient retrieval. Deep neural IRmodels, in contrast, compare the whole query to the document and are,therefore, typically employed only for late stage re-ranking. We incorporatequery term independence assumption into three state-of-the-art neural IRmodels: BERT, Duet, and CKNRM---and evaluate their performance on a passageranking task. Surprisingly, we observe no significant loss in result qualityfor Duet and CKNRM---and a small degradation in the case of BERT. However, byoperating on each query term independently, these otherwise computationallyintensive models become amenable to offline precomputation---dramaticallyreducing the cost of query evaluations employing state-of-the-art neuralranking models. This strategy makes it practical to use deep models forretrieval from large collections---and not restrict their usage to late stagere-ranking.

Incorporating Query Term Independence Assumption for Efficient Retrieval and Ranking using Deep Neural Networks