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

As the categories of named entities rapidly increase, the deployed NER modelsare required to keep updating toward recognizing more entity types, creating ademand for class-incremental learning for NER. Considering the privacy concernsand storage constraints, the standard paradigm for class-incremental NERupdates the models with training data only annotated with the new classes, yetthe entities from other entity classes are unlabeled, regarded as "Non-entity"(or "O"). In this work, we conduct an empirical study on the "Unlabeled EntityProblem" and find that it leads to severe confusion between "O" and entities,decreasing class discrimination of old classes and declining the model'sability to learn new classes. To solve the Unlabeled Entity Problem, we proposea novel representation learning method to learn discriminative representationsfor the entity classes and "O". Specifically, we propose an entity-awarecontrastive learning method that adaptively detects entity clusters in "O".Furthermore, we propose two effective distance-based relabeling strategies forbetter learning the old classes. We introduce a more realistic and challengingbenchmark for class-incremental NER, and the proposed method achieves up to10.62\% improvement over the baseline methods.

Learning "O" Helps for Learning More: Handling the Concealed Entity Problem for Class-incremental NER