“Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification
Jasmijn Bastings, Sebastian Ebert, Polina Zablotskaia, Anders Sandholm, Katja Filippova
Abstract
Feature attribution a.k.a. input salience methods which assign an importance score to a feature are abundant but may produce surprisingly different results for the same model on the same input. While differences are expected if disparate definitions of importance are assumed, most methods claim to provide faithful attributions and point at the features most relevant for a model’s prediction. Existing work on faithfulness evaluation is not conclusive and does not provide a clear answer as to how different methods are to be compared.Focusing on text classification and the model debugging scenario, our main contribution is a protocol for faithfulness evaluation that makes use of partially synthetic data to obtain ground truth for feature importance ranking. Following the protocol, we do an in-depth analysis of four standard salience method classes on a range of datasets and lexical shortcuts for BERT and LSTM models. We demonstrate that some of the most popular method configurations provide poor results even for simple shortcuts while a method judged to be too simplistic works remarkably well for BERT.- Anthology ID:
- 2022.emnlp-main.64
- Volume:
- Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Editors:
- Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 976–991
- Language:
- URL:
- https://aclanthology.org/2022.emnlp-main.64
- DOI:
- 10.18653/v1/2022.emnlp-main.64
- Cite (ACL):
- Jasmijn Bastings, Sebastian Ebert, Polina Zablotskaia, Anders Sandholm, and Katja Filippova. 2022. “Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 976–991, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- “Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification (Bastings et al., EMNLP 2022)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2022.emnlp-main.64.pdf