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

While there has been a recent explosion of work on ExplainableAI ExAI on deepmodels that operate on imagery and tabular data, textual datasets present newchallenges to the ExAI community. Such challenges can be attributed to the lackof input structure in textual data, the use of word embeddings that add to theopacity of the models and the difficulty of the visualization of the innerworkings of deep models when they are trained on textual data.  Lately, methods have been developed to address the aforementioned challengesand present satisfactory explanations on Natural Language Processing (NLP)models. However, such methods are yet to be studied in a comprehensiveframework where common challenges are properly stated and rigorous evaluationpractices and metrics are proposed. Motivated to democratize ExAI methods inthe NLP field, we present in this work a survey that studies model-agnostic aswell as model-specific explainability methods on NLP models. Such methods caneither develop inherently interpretable NLP models or operate on pre-trainedmodels in a post-hoc manner. We make this distinction and we further decomposethe methods into three categories according to what they explain: (1) wordembeddings (input-level), (2) inner workings of NLP models (processing-level)and (3) models' decisions (output-level). We also detail the differentevaluation approaches interpretability methods in the NLP field. Finally, wepresent a case-study on the well-known neural machine translation in anappendix and we propose promising future research directions for ExAI in theNLP field.

On the Explainability of Natural Language Processing Deep Models