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

Vocabulary selection, or lexical shortlisting, is a well-known technique toimprove latency of Neural Machine Translation models by constraining the set ofallowed output words during inference. The chosen set is typically determinedby separately trained alignment model parameters, independent of thesource-sentence context at inference time. While vocabulary selection appearscompetitive with respect to automatic quality metrics in prior work, we showthat it can fail to select the right set of output words, particularly forsemantically non-compositional linguistic phenomena such as idiomaticexpressions, leading to reduced translation quality as perceived by humans.Trading off latency for quality by increasing the size of the allowed set isoften not an option in real-world scenarios. We propose a model of vocabularyselection, integrated into the neural translation model, that predicts the setof allowed output words from contextualized encoder representations. Thisrestores translation quality of an unconstrained system, as measured by humanevaluations on WMT newstest2020 and idiomatic expressions, at an inferencelatency competitive with alignment-based selection using aggressive thresholds,thereby removing the dependency on separately trained alignment models.

The Devil is in the Details: On the Pitfalls of Vocabulary Selection in Neural Machine Translation