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

Debiasing methods in NLP models traditionally focus on isolating informationrelated to a sensitive attribute (e.g., gender or race). We instead argue thata favorable debiasing method should use sensitive information 'fairly,' withexplanations, rather than blindly eliminating it. This fair balance is oftensubjective and can be challenging to achieve algorithmically. We explore twointeractive setups with a frozen predictive model and show that users able toprovide feedback can achieve a better and fairer balance between taskperformance and bias mitigation. In one setup, users, by interacting with testexamples, further decreased bias in the explanations (5-8%) while maintainingthe same prediction accuracy. In the other setup, human feedback was able todisentangle associated bias and predictive information from the input leadingto superior bias mitigation and improved task performance (4-5%)simultaneously.

InterFair: Debiasing with Natural Language Feedback for Fair Interpretable Predictions