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

Machine learning models are prone to overfitting their training (source)domains, which is commonly believed to be the reason why they falter in noveltarget domains. Here we examine the contrasting view that multi-source domaingeneralization (DG) is first and foremost a problem of mitigating source domainunderfitting: models not adequately learning the signal already present intheir multi-domain training data. Experiments on a reading comprehension DGbenchmark show that as a model learns its source domains better -- usingfamiliar methods such as knowledge distillation (KD) from a bigger model -- itszero-shot out-of-domain utility improves at an even faster pace. Improvedsource domain learning also demonstrates superior out-of-domain generalizationover three popular existing DG approaches that aim to limit overfitting. Ourimplementation of KD-based domain generalization is available via PrimeQA at:https://ibm.biz/domain-generalization-with-kd.

Not to Overfit or Underfit the Source Domains? An Empirical Study of Domain Generalization in Question Answering