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

NLP models are used in a variety of critical social computing tasks, such asdetecting sexist, racist, or otherwise hateful content. Therefore, it isimperative that these models are robust to spurious features. Past work hasattempted to tackle such spurious features using training data augmentation,including Counterfactually Augmented Data (CADs). CADs introduce minimalchanges to existing training data points and flip their labels; training onthem may reduce model dependency on spurious features. However, manuallygenerating CADs can be time-consuming and expensive. Hence in this work, weassess if this task can be automated using generative NLP models. Weautomatically generate CADs using Polyjuice, ChatGPT, and Flan-T5, and evaluatetheir usefulness in improving model robustness compared to manually-generatedCADs. By testing both model performance on multiple out-of-domain test sets andindividual data point efficacy, our results show that while manual CADs arestill the most effective, CADs generated by ChatGPT come a close second. Onekey reason for the lower performance of automated methods is that the changesthey introduce are often insufficient to flip the original label.

People Make Better Edits: Measuring the Efficacy of LLM-Generated Counterfactually Augmented Data for Harmful Language Detection