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

Recent research on model interpretability in natural language processingextensively uses feature scoring methods for identifying which parts of theinput are the most important for a model to make a prediction (i.e. explanationor rationale). However, previous research has shown that there is no clear bestscoring method across various text classification tasks while practitionerstypically have to make several other ad-hoc choices regarding the length andthe type of the rationale (e.g. short or long, contiguous or not). Inspired bythis, we propose a simple yet effective and flexible method that allowsselecting optimally for each data instance: (1) a feature scoring method; (2)the length; and (3) the type of the rationale. Our method is inspired by inputerasure approaches to interpretability which assume that the most faithfulrationale for a prediction should be the one with the highest differencebetween the model's output distribution using the full text and the text afterremoving the rationale as input respectively. Evaluation on four standard textclassification datasets shows that our proposed method provides more faithful,comprehensive and highly sufficient explanations compared to using a fixedfeature scoring method, rationale length and type. More importantly, wedemonstrate that a practitioner is not required to make any ad-hoc choices inorder to extract faithful rationales using our approach.

Flexible Instance-Specific Rationalization of NLP Models