Beyond Accuracy: Behavioral Testing of NLP Models with Checklist (Extended Abstract)

Ribeiro, Marco Tulio; Wu, Tongshuang; Guestrin, Carlos; Singh, Sameer

doi:10.24963/ijcai.2021/659

Beyond Accuracy: Behavioral Testing of NLP Models with Checklist (Extended Abstract)

Marco Tulio Ribeiro, Tongshuang Wu, Carlos Guestrin, Sameer Singh

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Sister Conferences Best Papers. Pages 4824-4828.

https://doi.org/10.24963/ijcai.2021/659

Keywords

Natural Language Processing: Resources and Evaluation
Natural Language Processing: Question Answering
Natural Language Processing: Text Classification
Natural Language Processing: NLP Applications and Tools

Abstract

Although measuring held-out accuracy has been the primary approach to evaluate generalization, it often overestimates the performance of NLP models, while alternative approaches for evaluating models either focus on individual tasks or on specific behaviors. Inspired by principles of behavioral testing in software engineering, we introduce CheckList, a task-agnostic methodology for testing NLP models. CheckList includes a matrix of general linguistic capabilities and test types that facilitate comprehensive test ideation, as well as a software tool to generate a large and diverse number of test cases quickly. We illustrate the utility of CheckList with tests for three tasks, identifying critical failures in both commercial and state-of-art models. In a user study, a team responsible for a commercial sentiment analysis model found new and actionable bugs in an extensively tested model. In another user study, NLP practitioners with CheckList created twice as many tests, and found almost three times as many bugs as users without it.

Marco Tulio Ribeiro, Tongshuang Wu, Carlos Guestrin, and Sameer Singh. 2021. Beyond Accuracy: Behavioral Testing of NLP Models with Checklist (Extended Abstract). In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 4824–4828. https://doi.org/10.24963/ijcai.2021/659

@inproceedings{ijcai2021p659,
  title     = {Beyond Accuracy: Behavioral Testing of NLP Models with Checklist (Extended Abstract)},
  author    = {Ribeiro, Marco Tulio and Wu, Tongshuang and Guestrin, Carlos and Singh, Sameer},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Prof. Zhi-Hua Zhou},
  pages     = {4824--4828},
  year      = {2021},
  month     = {8},
  note      = {Sister Conferences Best Papers},
  doi       = {10.24963/ijcai.2021/659},
  url       = {https://doi.org/10.24963/ijcai.2021/659},
}

TY  - CPAPER
TI  - Beyond Accuracy: Behavioral Testing of NLP Models with Checklist (Extended Abstract)
AU  - Ribeiro, Marco Tulio
AU  - Wu, Tongshuang
AU  - Guestrin, Carlos
AU  - Singh, Sameer
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 4824
EP  - 4828
DO  - 10.24963/ijcai.2021/659
UR  - https://doi.org/10.24963/ijcai.2021/659
ER  -