Towards Faithful Model Explanation in NLP: A Survey

Qing Lyu, Marianna Apidianaki, Chris Callison-Burch


Abstract
End-to-end neural Natural Language Processing (NLP) models are notoriously difficult to understand. This has given rise to numerous efforts towards model explainability in recent years. One desideratum of model explanation is faithfulness, that is, an explanation should accurately represent the reasoning process behind the model’s prediction. In this survey, we review over 110 model explanation methods in NLP through the lens of faithfulness. We first discuss the definition and evaluation of faithfulness, as well as its significance for explainability. We then introduce recent advances in faithful explanation, grouping existing approaches into five categories: similarity-based methods, analysis of model-internal structures, backpropagation-based methods, counterfactual intervention, and self-explanatory models. For each category, we synthesize its representative studies, strengths, and weaknesses. Finally, we summarize their common virtues and remaining challenges, and reflect on future work directions towards faithful explainability in NLP.
Anthology ID:
2024.cl-2.6
Volume:
Computational Linguistics, Volume 50, Issue 2 - June 2023
Month:
June
Year:
2024
Address:
Cambridge, MA
Venue:
CL
SIG:
Publisher:
MIT Press
Note:
Pages:
657–723
Language:
URL:
https://aclanthology.org/2024.cl-2.6
DOI:
10.1162/coli_a_00511
Bibkey:
Cite (ACL):
Qing Lyu, Marianna Apidianaki, and Chris Callison-Burch. 2024. Towards Faithful Model Explanation in NLP: A Survey. Computational Linguistics, 50(2):657–723.
Cite (Informal):
Towards Faithful Model Explanation in NLP: A Survey (Lyu et al., CL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.cl-2.6.pdf