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 success of large-scale pre-trained language models crucially hinge onfine-tuning them on large amounts of labeled data for the downstream task, thatare typically expensive to acquire. In this work, we study self-training as oneof the earliest semi-supervised learning approaches to reduce the annotationbottleneck by making use of large-scale unlabeled data for the target task.Standard self-training mechanism randomly samples instances from the unlabeledpool to pseudo-label and augment labeled data. In this work, we propose anapproach to improve self-training by incorporating uncertainty estimates of theunderlying neural network leveraging recent advances in Bayesian deep learning.Specifically, we propose (i) acquisition functions to select instances from theunlabeled pool leveraging Monte Carlo (MC) Dropout, and (ii) learning mechanismleveraging model confidence for self-training. As an application, we focus ontext classification on five benchmark datasets. We show our methods leveragingonly 20-30 labeled samples per class for each task for training and forvalidation can perform within 3% of fully supervised pre-trained languagemodels fine-tuned on thousands of labeled instances with an aggregate accuracyof 91% and improving by upto 12% over baselines.

Uncertainty-aware Self-training for Text Classification with Few Labels