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

Recent advances in NLP demonstrate the effectiveness of training large-scalelanguage models and transferring them to downstream tasks. Can fine-tuningthese models on tasks other than language modeling further improve performance?In this paper, we conduct an extensive study of the transferability between 33NLP tasks across three broad classes of problems (text classification, questionanswering, and sequence labeling). Our results show that transfer learning ismore beneficial than previously thought, especially when target task data isscarce, and can improve performance even when the source task is small ordiffers substantially from the target task (e.g., part-of-speech taggingtransfers well to the DROP QA dataset). We also develop task embeddings thatcan be used to predict the most transferable source tasks for a given targettask, and we validate their effectiveness in experiments controlled for sourceand target data size. Overall, our experiments reveal that factors such assource data size, task and domain similarity, and task complexity all play arole in determining transferability.

Exploring and Predicting Transferability across NLP Tasks