[CheckList] Beyond Accuracy: Behavioral Testing of NLP Models with CheckList

Time: Thursday 27-Aug-2020 16:00 (This is a past event.)

Discussion Facilitator:

Artifacts

Motivation / Abstract
- The paper proposes CheckList, a novel behavioural testing methodology
- CheckList provides you tools that will  you build software engineering like test cases at scale!
- Using CheckList, the paper identifies critical failures in both commercial
and state-of-the-art models
- This paper won the overall Best Paper Award at ACL'20
Stream Categories:
 Natural Language Processing