On the Impact of Temporal Concept Drift on Model Explanations

Zhixue Zhao, George Chrysostomou, Kalina Bontcheva, Nikolaos Aletras

Abstract
Explanation faithfulness of model predictions in natural language processing is typically evaluated on held-out data from the same temporal distribution as the training data (i.e. synchronous settings). While model performance often deteriorates due to temporal variation (i.e. temporal concept drift), it is currently unknown how explanation faithfulness is impacted when the time span of the target data is different from the data used to train the model (i.e. asynchronous settings). For this purpose, we examine the impact of temporal variation on model explanations extracted by eight feature attribution methods and three select-then-predict models across six text classification tasks. Our experiments show that (i) faithfulness is not consistent under temporal variations across feature attribution methods (e.g. it decreases or increases depending on the method), with an attention-based method demonstrating the most robust faithfulness scores across datasets; and (ii) select-then-predict models are mostly robust in asynchronous settings with only small degradation in predictive performance. Finally, feature attribution methods show conflicting behavior when used in FRESH (i.e. a select-and-predict model) and for measuring sufficiency/comprehensiveness (i.e. as post-hoc methods), suggesting that we need more robust metrics to evaluate post-hoc explanation faithfulness. Code will be made publicly available.
Anthology ID:
2022.findings-emnlp.298
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4039–4054
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.298
DOI:
10.18653/v1/2022.findings-emnlp.298
Bibkey:
Cite (ACL):
Zhixue Zhao, George Chrysostomou, Kalina Bontcheva, and Nikolaos Aletras. 2022. On the Impact of Temporal Concept Drift on Model Explanations. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 4039–4054, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
On the Impact of Temporal Concept Drift on Model Explanations (Zhao et al., Findings 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-2024-clasp/2022.findings-emnlp.298.pdf
PDF Search