Abstract
Counterfactuals are a valuable means for understanding decisions made by ML systems. However, the counterfactuals generated by the methods currently available for natural language text are either unrealistic or introduce imperceptible changes. We propose CounterfactualGAN: a method that combines a conditional GAN and the embeddings of a pretrained BERT encoder to model-agnostically generate realistic natural language text counterfactuals for explaining regression and classification tasks. Experimental results show that our method produces perceptibly distinguishable counterfactuals, while outperforming four baseline methods on fidelity and human judgments of naturalness, across multiple datasets and multiple predictive models.- Anthology ID:
- 2021.findings-emnlp.306
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2021
- Month:
- November
- Year:
- 2021
- Address:
- Punta Cana, Dominican Republic
- Venue:
- Findings
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 3611–3625
- Language:
- URL:
- https://aclanthology.org/2021.findings-emnlp.306
- DOI:
- 10.18653/v1/2021.findings-emnlp.306
- Cite (ACL):
- Marcel Robeer, Floris Bex, and Ad Feelders. 2021. Generating Realistic Natural Language Counterfactuals. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 3611–3625, Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Generating Realistic Natural Language Counterfactuals (Robeer et al., Findings 2021)
- PDF:
- https://preview.aclanthology.org/paclic-22-ingestion/2021.findings-emnlp.306.pdf
- Data
- SNLI, SST