Evaluating and Characterizing Human Rationales

Samuel Carton, Anirudh Rathore, Chenhao Tan


Abstract
Two main approaches for evaluating the quality of machine-generated rationales are: 1) using human rationales as a gold standard; and 2) automated metrics based on how rationales affect model behavior. An open question, however, is how human rationales fare with these automatic metrics. Analyzing a variety of datasets and models, we find that human rationales do not necessarily perform well on these metrics. To unpack this finding, we propose improved metrics to account for model-dependent baseline performance. We then propose two methods to further characterize rationale quality, one based on model retraining and one on using “fidelity curves” to reveal properties such as irrelevance and redundancy. Our work leads to actionable suggestions for evaluating and characterizing rationales.
Anthology ID:
2020.emnlp-main.747
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Editors:
Bonnie Webber, Trevor Cohn, Yulan He, Yang Liu
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
9294–9307
Language:
URL:
https://aclanthology.org/2020.emnlp-main.747
DOI:
10.18653/v1/2020.emnlp-main.747
Bibkey:
Cite (ACL):
Samuel Carton, Anirudh Rathore, and Chenhao Tan. 2020. Evaluating and Characterizing Human Rationales. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 9294–9307, Online. Association for Computational Linguistics.
Cite (Informal):
Evaluating and Characterizing Human Rationales (Carton et al., EMNLP 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.emnlp-main.747.pdf
Video:
 https://slideslive.com/38939202
Code
 BoulderDS/evaluating-human-rationales
Data
FEVERMultiRCSSTe-SNLI