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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Named entity recognition (NER) is a fundamental component in the modernlanguage understanding pipeline. Public NER resources such as annotated dataand model services are available in many domains. However, given a particulardownstream application, there is often no single NER resource that supports allthe desired entity types, so users must leverage multiple resources withdifferent tag sets. This paper presents a marginal distillation (MARDI)approach for training a unified NER model from resources with disjoint orheterogeneous tag sets. In contrast to recent works, MARDI merely requiresaccess to pre-trained models rather than the original training datasets. Thisflexibility makes it easier to work with sensitive domains like healthcare andfinance. Furthermore, our approach is general enough to integrate withdifferent NER architectures, including local models (e.g., BiLSTM) and globalmodels (e.g., CRF). Experiments on two benchmark datasets show that MARDIperforms on par with a strong marginal CRF baseline, while being more flexiblein the form of required NER resources. MARDI also sets a new state of the arton the progressive NER task. MARDI significantly outperforms thestart-of-the-art model on the task of progressive NER.

One Model to Recognize Them All: Marginal Distillation from NER Models with Different Tag Sets