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

We propose a genetic algorithm (GA) based method for modifying n-best listsproduced by a machine translation (MT) system. Our method offers an innovativeapproach to improving MT quality and identifying weaknesses in evaluationmetrics. Using common GA operations (mutation and crossover) on a list ofhypotheses in combination with a fitness function (an arbitrary MT metric), weobtain novel and diverse outputs with high metric scores. With a combination ofmultiple MT metrics as the fitness function, the proposed method leads to anincrease in translation quality as measured by other held-out automaticmetrics. With a single metric (including popular ones such as COMET) as thefitness function, we find blind spots and flaws in the metric. This allows foran automated search for adversarial examples in an arbitrary metric, withoutprior assumptions on the form of such example. As a demonstration of themethod, we create datasets of adversarial examples and use them to show thatreference-free COMET is substantially less robust than the reference-basedversion.

Breeding Machine Translations: Evolutionary approach to survive and thrive in the world of automated evaluation