Parallel Sentence-Level Explanation Generation for Real-World Low-Resource Scenarios
This work addresses the need for efficient and accessible interpretability in real-world applications where annotated data is scarce, though it is incremental in advancing from weak-supervised to unsupervised methods.
The paper tackles the problem of generating sentence-level explanations for model predictions in low-resource scenarios, achieving a 10-15x faster training speed for classifiers with comparable performance using few or no annotated data.
In order to reveal the rationale behind model predictions, many works have exploited providing explanations in various forms. Recently, to further guarantee readability, more and more works turn to generate sentence-level human language explanations. However, current works pursuing sentence-level explanations rely heavily on annotated training data, which limits the development of interpretability to only a few tasks. As far as we know, this paper is the first to explore this problem smoothly from weak-supervised learning to unsupervised learning. Besides, we also notice the high latency of autoregressive sentence-level explanation generation, which leads to asynchronous interpretability after prediction. Therefore, we propose a non-autoregressive interpretable model to facilitate parallel explanation generation and simultaneous prediction. Through extensive experiments on Natural Language Inference task and Spouse Prediction task, we find that users are able to train classifiers with comparable performance $10-15\times$ faster with parallel explanation generation using only a few or no annotated training data.