Catch up on the latest AI articles

What Is Explainability And A Thorough Explanation Of The Current State Of Explainable AI In NLP!

What Is Explainability And A Thorough Explanation Of The Current State Of Explainable AI In NLP!


3 main points
✔️ On Explainable AI in Natural Language Processing 
✔️ On the current state of explainable natural language processing models 
✔️ The challenges of achieving accountability

A Survey of the State of Explainable AI for Natural Language Processing
written by Marina DanilevskyKun QianRanit AharonovYannis KatsisBan KawasPrithviraj Sen
(Submitted on 1 Oct 2020)

Comments: Accepted at AACL-IJCNLP2020
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)


Deep learning models, which have enjoyed great success in recent years, face the problem of difficulty in interpreting the criteria for their models (they are black boxes).

This lack of accountability can also undermine the credibility of the model with users.

For this reason, research on Explainable AI (XAI ), which reveals the reasons/rationale for decisions of deep learning models, has been gaining importance in recent years. Although this research on Explainable AI is a very important subject, it is a new area of research and therefore difficult to gain systematic knowledge.

In light of this lack of information, this article summarizes the current state of Explainable AI in natural language processing. If you are interested in Explainable AI, please take a look at this article.

table of contents

1. Broad Classification of Explainable AI 

2. details of the technology used for explainable AI 
 2.1. Explainability techniques 
  Feature importance 
  Surrogate model 
  Declarative induction

 2.2. Relevant Technologies of Accountability 
  First-derivative saliency 
  Layer-wise relevance propagation 
  Input perturbations 
  LSTM gating signals 
  Explainability-aware architecture design

 2.3. Explainability Visualization Techniques 
  Raw declarative representations 
  Natural language explanation

3. Assessing Explainability 
 3.1. Classification of the evaluation 
  An informal examination of explanations 
  Comparing to ground truth 
  Human evaluation 

 3.2. what is explicable?

4. Summary of actual research cases

To read more,

Please register with AI-SCHOLAR.

Sign up for free in 1 minute


If you have any suggestions for improvement of the content of the article,
please contact the AI-SCHOLAR editorial team through the contact form.

Contact Us