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

Existing NLP datasets contain various biases, and models tend to quicklylearn those biases, which in turn limits their robustness. Existing approachesto improve robustness against dataset biases mostly focus on changing thetraining objective so that models learn less from biased examples. Besides,they mostly focus on addressing a specific bias, and while they improve theperformance on adversarial evaluation sets of the targeted bias, they may biasthe model in other ways, and therefore, hurt the overall robustness. In thispaper, we propose to augment the input sentences in the training data withtheir corresponding predicate-argument structures, which provide a higher-levelabstraction over different realizations of the same meaning and help the modelto recognize important parts of sentences. We show that without targeting aspecific bias, our sentence augmentation improves the robustness of transformermodels against multiple biases. In addition, we show that models can still bevulnerable to the lexical overlap bias, even when the training data does notcontain this bias, and that the sentence augmentation also improves therobustness in this scenario. We will release our adversarial datasets toevaluate bias in such a scenario as well as our augmentation scripts athttps://github.com/UKPLab/data-augmentation-for-robustness.

Improving Robustness by Augmenting Training Sentences with Predicate-Argument Structures