On Completeness-aware Concept-Based Explanations in Deep Neural Networks

Meta Review

I recommend accepting this paper. I believe that this work will be of relevance to the machine learning community. Although reviewers initially had some doubts about a few aspects of the methodology, they were convincingly addressed in a rebuttal.