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

Recent advances in retrieval models based on learned sparse representationsgenerated by transformers have led us to, once again, consider score-at-a-timequery evaluation techniques for the top-k retrieval problem. Previous studiescomparing document-at-a-time and score-at-a-time approaches have consistentlyfound that the former approach yields lower mean query latency, although thelatter approach has more predictable query latency. In our experiments withfour different retrieval models that exploit representational learning withbags of words, we find that transformers generate "wacky weights" that appearto greatly reduce the opportunities for skipping and early exitingoptimizations that lie at the core of standard document-at-a-time techniques.As a result, score-at-a-time approaches appear to be more competitive in termsof query evaluation latency than in previous studies. We find that, if aneffectiveness loss of up to three percent can be tolerated, a score-at-a-timeapproach can yield substantial gains in mean query latency while at the sametime dramatically reducing tail latency.

Wacky Weights in Learned Sparse Representations and the Revenge of Score-at-a-Time Query Evaluation