Making a (Counterfactual) Difference One Rationale at a Time
This addresses the issue of nonsensical behaviors in interpretable NLP models for researchers and practitioners, representing an incremental improvement.
The paper tackled the problem of spurious text patterns in rationale models for interpretable NLP by using unsupervised counterfactual data augmentation, showing that it produces rationales that better capture the signal of interest compared to baselines on two multi-aspect datasets.
Rationales, snippets of extracted text that explain an inference, have emerged as a popular framework for interpretable natural language processing (NLP). Rationale models typically consist of two cooperating modules: a selector and a classifier with the goal of maximizing the mutual information (MMI) between the "selected" text and the document label. Despite their promises, MMI-based methods often pick up on spurious text patterns and result in models with nonsensical behaviors. In this work, we investigate whether counterfactual data augmentation (CDA), without human assistance, can improve the performance of the selector by lowering the mutual information between spurious signals and the document label. Our counterfactuals are produced in an unsupervised fashion using class-dependent generative models. From an information theoretic lens, we derive properties of the unaugmented dataset for which our CDA approach would succeed. The effectiveness of CDA is empirically evaluated by comparing against several baselines including an improved MMI-based rationale schema on two multi aspect datasets. Our results show that CDA produces rationales that better capture the signal of interest.