Goodhart’s Law Applies to NLP’s Explanation Benchmarks

Jennifer Hsia, Danish Pruthi, Aarti Singh, Zachary Lipton


Abstract
Despite the rising popularity of saliency-based explanations, the research community remains at an impasse, facing doubts concerning their purpose, efficacy, and tendency to contradict each other. Seeking to unite the community’s efforts around common goals, several recent works have proposed evaluation metrics. In this paper, we critically examine two sets of metrics: the ERASER metrics (comprehensiveness and sufficiency) and the EVAL-X metrics, focusing our inquiry on natural language processing. First, we show that we can inflate a model’s comprehensiveness and sufficiency scores dramatically without altering its predictions or explanations on in-distribution test inputs. Our strategy exploits the tendency for extracted explanations and their complements to be “out-of-support” relative to each other and in-distribution inputs. Next, we demonstrate that the EVAL-X metrics can be inflated arbitrarily by a simple method that encodes the label, even though EVAL-X is precisely motivated to address such exploits. Our results raise doubts about the ability of current metrics to guide explainability research, underscoring the need for a broader reassessment of what precisely these metrics are intended to capture.
Anthology ID:
2024.findings-eacl.88
Volume:
Findings of the Association for Computational Linguistics: EACL 2024
Month:
March
Year:
2024
Address:
St. Julian’s, Malta
Editors:
Yvette Graham, Matthew Purver
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1322–1335
Language:
URL:
https://aclanthology.org/2024.findings-eacl.88
DOI:
Bibkey:
Cite (ACL):
Jennifer Hsia, Danish Pruthi, Aarti Singh, and Zachary Lipton. 2024. Goodhart’s Law Applies to NLP’s Explanation Benchmarks. In Findings of the Association for Computational Linguistics: EACL 2024, pages 1322–1335, St. Julian’s, Malta. Association for Computational Linguistics.
Cite (Informal):
Goodhart’s Law Applies to NLP’s Explanation Benchmarks (Hsia et al., Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-1/2024.findings-eacl.88.pdf