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 explainability related studies have shown that state-of-the-art DNNsdo not always adopt correct evidences to make decisions. It not only hamperstheir generalization but also makes them less likely to be trusted byend-users. In pursuit of developing more credible DNNs, in this paper wepropose CREX, which encourages DNN models to focus more on evidences thatactually matter for the task at hand, and to avoid overfitting todata-dependent bias and artifacts. Specifically, CREX regularizes the trainingprocess of DNNs with rationales, i.e., a subset of features highlighted bydomain experts as justifications for predictions, to enforce DNNs to generatelocal explanations that conform with expert rationales. Even when rationalesare not available, CREX still could be useful by requiring the generatedexplanations to be sparse. Experimental results on two text classificationdatasets demonstrate the increased credibility of DNNs trained with CREX.Comprehensive analysis further shows that while CREX does not always improveprediction accuracy on the held-out test set, it significantly increases DNNaccuracy on new and previously unseen data beyond test set, highlighting theadvantage of the increased credibility.

Learning Credible Deep Neural Networks with Rationale Regularization