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

Representation learning is a critical ingredient for natural languageprocessing systems. Recent Transformer language models like BERT learn powerfultextual representations, but these models are targeted towards token- andsentence-level training objectives and do not leverage information oninter-document relatedness, which limits their document-level representationpower. For applications on scientific documents, such as classification andrecommendation, the embeddings power strong performance on end tasks. Wepropose SPECTER, a new method to generate document-level embedding ofscientific documents based on pretraining a Transformer language model on apowerful signal of document-level relatedness: the citation graph. Unlikeexisting pretrained language models, SPECTER can be easily applied todownstream applications without task-specific fine-tuning. Additionally, toencourage further research on document-level models, we introduce SciDocs, anew evaluation benchmark consisting of seven document-level tasks ranging fromcitation prediction, to document classification and recommendation. We showthat SPECTER outperforms a variety of competitive baselines on the benchmark.

SPECTER: Document-level Representation Learning using Citation-informed Transformers