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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Due to high annotation costs making the best use of existing human-createdtraining data is an important research direction. We, therefore, carry out asystematic evaluation of transferability of BERT-based neural ranking modelsacross five English datasets. Previous studies focused primarily on zero-shotand few-shot transfer from a large dataset to a dataset with a small number ofqueries. In contrast, each of our collections has a substantial number ofqueries, which enables a full-shot evaluation mode and improves reliability ofour results. Furthermore, since source datasets licences often prohibitcommercial use, we compare transfer learning to training on pseudo-labelsgenerated by a BM25 scorer. We find that training on pseudo-labels -- possiblywith subsequent fine-tuning using a modest number of annotated queries -- canproduce a competitive or better model compared to transfer learning. Yet, it isnecessary to improve the stability and/or effectiveness of the few-shottraining, which, sometimes, can degrade performance of a pretrained model.

A Systematic Evaluation of Transfer Learning and Pseudo-labeling with BERT-based Ranking Models